Field trip to the Mesoarchaean Barberton Granite-Greenstone

Transcription

Sixth International Hutton Symposium on Granites and Related Rocks
2–6 July 2007, Stellenbosch, South Africa
FT2 Post-conference
Field trip to the Mesoarchaean Barberton
Granite-Greenstone Terrain
Jean-François Moyen
Alexander F.M. Kisters
Gary Stevens
With the participation of Christiano Lana and Richard W. Belcher
June 2007
University of Stellenbosch, South Africa
∗
Cover photo: Top, view from the “Horseshoe hill” in Inyoni Shear Zone, looking North towards
the Stolzburg pluton and the BGB. Bottom left, ca. 3.2 Ga shear zones in the 3.45 Ga Stolzburg pluton,
near the granite–greenstone contact. Middle, Lit-par-lit intrusion of the Stolzburg pluton in a greenstone
remnant, Theespruit River. Right, a dyke of ca. 3.1 Ga Boesmanskop syenite cutting banded gneisses
in the Welverdiend Shear Zone (stop 3.6)
iii
Abbreviated program
• Saturday, 7 July 2007: Fly from Cape Town to Johannesburg; drive to Badplaas.
• Sunday, 8 July 2007: The lower part of Barberton Greenstone belt, and the granite–
greenstone contacts in the ca. 3.55–3.45 Ga domain.
• Monday, 9 July 2007: Structures and metamorphism in the ca. 3.45 Ga Stolzburg
terrane; the ca. 3.1 Ga Boesmanskop syenogranitic complex.
• Tuesday, 10 July 2007: The ca. 3.2 Ga plutons on the North-Eastern side of the
Barberton Greenstone Belt.
• Wednesday, 11 July 2007: The ca. 3.1 Ga granitic plutons in the South of the Barberton
Granite–Greenstone Terrain.
• Thursday, 12 July 2007: Drive back to Johannesburg.
Accommodation
Forever Resorts – Aventura Badplaas
Tel: +27 (0) 17 844 8000
Fax: +27 (0) 17 844 1391
E-mail: [email protected]
iv
25°45'0"S
26°0'0"S
Machadodorp
Carolina
26°15'0"S
Carolina
5.1
5.2
30°30'0"E
5.4
Heerenveen
Transvaal Supergroup
4.1
Rhg.
5.3
Warburton
Karoo Supergroup
30°30'0"E
5.5
3.5
3.4
4.2
3.7
3.6
BMK
Stz.
Nelshoogte
4.3
Ko
4.4
3.3
0 2.5 5
Nelspruit
10
31°0'0"E
15
Barberton
20
Kilometers
4.6
4.7
Mbabane
Pigg's
Peak
Josefsdal
4.5
eKulindeni
2.1
31°0'0"E
Oshoek
Sty.
2.2
Kom
ati
Barberton Greenstone Belt
Kaap Valley
Sty: Steynsdorp
Ths: Theespruit
Stz: Stolzburg
3.2
2.3
Sw
A . nd
S. ila
R. az
Dalmein
Tjakastad
Ths.
2.4
Lochiel
eLukwatini
3.1
ma
ti
30°45'0"E
BMK: Boesmanskop Bpl: Badplaas
Rhg: Rooihoogte
Badplaas
Bpl.
t
rui
sp
ee
Th
Schapenburg
Mpuluzi
30°45'0"E
25°45'0"S
26°0'0"S
26°15'0"S
University
University of Missouri
Geoscience Australia
Dionyz Stur State Institute of Geology
Geoscience Research Institute
University of Geneva
University of Washington Seattle
University of Missouri
INGEIS
Universität Stuttgart
University of Puerto Rico
Geol. Surv. NSW/ Monash University
Geological Survey of Japan
Geological Survey of Canada
Ben Gurion University of the Negev
Ben Gurion University of the Negev
Université Toulouse III
James Cook University
University of Ottawa
The Australian National University
Université Clermont–Ferrand II
Council for Geoscience, South Africa
University of Stellenbosch
Surname
Nabelek
Champion
Kohut
Clausen
Annen
Evans
Whittington
Lopez de Luchi
Massonne
Cavosie
Quinn
Nakajima
Bédard
Katzir
Be’eri-Shlevin
Nédélec
Collins
Benn
Rapp
Martin
Belcher
Stevens
Moyen
Kisters
Sanchez–Garrido
Taylor
Name
Peter
David
Milan
Ben
Catherine
Bernard
Alan
Monica
Hans-Joachim
Aaron
Cameron
Takashi
Jean
Yaron
Yaron
Anne
William J.
Keith
Robert P.
Hervé
Richard
Gary
Jean-François
Alexander
Cynthia
Jeanne
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
Email
Excursion route
v
Field trip participants
vi
Contents
Abbreviated program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Excursion route . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
List of participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I
Introduction to the geology of the Barberton terrain
1.
2.
3.
4.
5.
iii
iv
v
1
Stratigraphy of the Barberton Greenstone Belt . . . . . . . . .
Main tectono-magmatic events . . . . . . . . . . . . . . . . . .
Accretion stage of the BGGT at > 3.42 Ga (pre-D1 and D1 ) . .
3.1.
An oceanic plateau. . . . . . . . . . . . . . . . . . . . . .
3.2.
. . . modified by a subduction (?) event . . . . . . . . . .
3.3.
Renewed (ultra)mafic activity . . . . . . . . . . . . . . .
A subduction-collision-exhumation orogen at 3.29-3.21 Ga (D2 )
4.1.
Evidence for accretionary orogen in the BGGT . . . . .
4.2.
Metamorphism associated with the ca. 3.2 Ga orogen .
4.3.
Ca. 3.2 Ga magmatism . . . . . . . . . . . . . . . . . .
4.4.
Evolution model . . . . . . . . . . . . . . . . . . . . . .
The sheeted batholiths of the GMS suite (3.11 Ga) . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
6
7
7
7
9
9
9
12
12
14
16
II Field itinerary
21
Saturday, 7 July
23
Sunday, 8 July
Komati river section, Songimvelo nature reserve . . .
Contact of the Steynsdorp pluton . . . . . . . . . . .
Contact of the Dalmein pluton . . . . . . . . . . . .
Deformed intrusive breccia of the Theespruit pluton
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
25
25
28
31
31
Monday, 9 July
Stolzburg pluton contact . . . . . . . . . . . . .
Felsic agglomerates in the Tjakastad schist belt
Komatiite type locality . . . . . . . . . . . . . .
The ca. 3.2 Ga Inyoni shear Zone (ISZ) . . . .
Western slopes of Boesmanskop . . . . . . . . .
Basal (?) contact of Boesmanskop pluton . . .
Hypovolcanic facies of the Boesmanskop syenite
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
33
34
36
36
40
41
41
. . .
. . .
. . .
. . .
town
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43
43
44
46
46
47
Tuesday, 10 July
Rooihoogte pluton . . . . . . . .
The Nelshoogte pluton . . . . . .
Nelshoogte pass . . . . . . . . . .
Northern side of Nelshoogte pass
Border of the Kaap Valley pluton
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
above Barberton
vii
.
.
.
.
.
.
.
viii
Contents
Moodies conglomerates along the R40 . . . . . . . . . . . . . . . . . . . . . . . . . . .
Panorama on the Fig Tree Valley . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Wednesday, 11 July
Pavements of the ca. 3.1 Ga Heerenveen granite . . .
Intrusive breccias in the Heerenveen batholith . . . . .
Synmagmatic shear zones in the Heerenveen batholith
Schapenburg Greenstone remnant . . . . . . . . . . . .
Eagle Heights . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Thursday, 12 July
51
51
52
53
55
58
61
III Articles
The 3.2 Ga orogeny: structures and metamorphism
1.
Kisters et al., 2003 . . . . . . . . . . .
2.
Stevens and Moyen, in press . . . . . .
3.
Moyen et al., 2006 . . . . . . . . . . .
4.
Diener et al., 2005 . . . . . . . . . . .
Petrology and geochemistry of the TTG suite . . .
5.
Clemens et al., 2006 . . . . . . . . . .
6.
Moyen et al., in press . . . . . . . . .
Emplacement of the ca. 3.1 Ga GMS suite . . . . .
7.
Belcher and Kisters, 2006 . . . . . . .
8.
Westraat et al., 2005 . . . . . . . . . .
47
48
63
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
65
66
78
93
95
108
108
121
152
152
160
List of Figures
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Simplified geological map of the Barberton Greenstone belt . . . . . . . .
Simplified log in the Barberton Greenstone Belt . . . . . . . . . . . . . . .
3.55-3.35 Ga history of the BGB . . . . . . . . . . . . . . . . . . . . . . .
Intrusive relations of the Theespruit pluton with surrounding greenstones
Distinct terranes in the BGB . . . . . . . . . . . . . . . . . . . . . . . . .
Synthetic logs in the different terranes . . . . . . . . . . . . . . . . . . . .
Delineation of tectonic terranes in the BGB . . . . . . . . . . . . . . . . .
Emplacement model for the Nelshoogte pluton . . . . . . . . . . . . . . .
Geodynamical model for the ca. 3.2 Ga evolution of the BGGT . . . . . .
Extend of the ca. 3.1 Ga GMS suite . . . . . . . . . . . . . . . . . . . . .
Road map, from Johannesburg to Badplaas . . . . . . . . . . . . . . . . .
Geological map of the komati valley in Songimvelo Nature Reserve . . . .
Log in the Onvewacht group . . . . . . . . . . . . . . . . . . . . . . . . . .
Geological map of the Steynsdorp pluton and surroundings . . . . . . . .
Geological and structural map of the core of the Steynsdorp anticline . . .
Geological map of the Tjakastad schist belt and its contacts . . . . . . . .
Geological map of the Inyoni Shear Zone . . . . . . . . . . . . . . . . . . .
Metamorphic textures in Inyoni Shear Zone amphibolites . . . . . . . . .
Summary of P–T estimates from the Inyoni Shear Zone . . . . . . . . . .
Geological map of the Nelshoogte pluton . . . . . . . . . . . . . . . . . . .
Strain pattern and intensity in the Nelshoogte pluton . . . . . . . . . . .
Geological and structural map of the Heerenveen batholith . . . . . . . .
Geological map of the Schapenburg Greenstone remnant . . . . . . . . . .
Geological map of the North-Western edge of the Mpuluzi batholith . . .
Conceptualized map of the outcrops aroud locality 5.5 . . . . . . . . . . .
ix
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
5
7
8
10
10
11
13
15
16
23
26
27
28
30
35
37
39
39
44
45
53
55
58
59
Part I
Introduction to the geology of the
Barberton Granite–Greenstone terrain
he meso-Archaean Barberton granitegreenstone terrain (BGGT) consists of
two main components: the Barberton Greenstone Belt (BGB) proper, and the surrounding granitoids. The granitoids can further be
subdivided into an early trondhjemite-tonalitegranodiorite (TTG) suite, that formed before
or synchronous with the lower greenstone belt
stratigraphy, and a younger, post-greenstone,
granite-monzonite-syenite (GMS) suite. Together, the belt and the TTGs define a regional
dome-and-keel geometry, typical of many Archaean provinces.
T
Some recent papers from our group are reproduced in the appendix. They provide all the
details and the data supporting the models discussed below, such that only an overview is
given here.
U3
£+
O9
o
u~
c~
I
o.\
co
o
03
,08008
Oe
-'=-44
c~o
Zme3
0
I l l :D ¢/J 03
I.IiI ~ 0O ¢J3
,-,
"~=
-
,~..~
A n Archaean arc-arc collisional e v e n t
a<
~,.../~
~:~
/-"
Zh'm---o,.
-'''~o©
~
r- r.,D
~-
\
\
.
~-~
'
'
'-
]
n
E
~,-
\
~
.
--
"~
U~ q= v
"~
=_,
~=
0
E
v
O 04
O
11i-I
"°"
O
> o ~
.--
,-a
"6
=~-
~s
I:1.
~ +, "..\
(,/'J I:},,. e3
m..T.
- .~ {~ ~ _ ~
~
_ _
-
-
-
®
Journal of African Earth Sciences 221
Figure 1: Simplified geological map of the Barberton Greenstone belt and surrounding granite-gneiss terrain
(De Ronde and Kamo, 2000).
3
4
Introduction to the geology of the BGGT
1. Stratigraphy of the Barberton Greenstone Belt
An apparently simple stratigraphy has been
proposed for the BGB (figure 2); three groups
(collectively forming the Swaziland Supergroup) have been recognized, from the base upwards (Anhaeusser, 1969; Viljoen and Viljoen,
1969b; Visser, 1956):
The BGB essentially forms a complex synform.
The outermost levels, in contact with the TTG
plutons generally correspond to the lowest
stratigraphic levels, and are typically metamorphosed to amphibolite facies grades, whereas
the core of the belt records mid- to lowergreenschist facies condition.
Amphibolitefacies
rocks
(from
the
BGB,
and
nearby TTG
1. The Onverwacht group (ca.
3.5—
orthogneisses)
form
a
prominent
fault-bounded
3.3 Ga) consists predominantly of maficultramafic volcanic rocks, including ko- region South of the BGB, that has been termed
matiites; minor intercalations of chert the “Stolzburg domain” (Dziggel et al., 2002;
bands, felsic lavas and clastic sediments Kisters et al., 2003; Moyen et al., 2006; Stevens
(De Wit et al., 1987; Lowe and Byerly, and Moyen, in press).
1999) are also found.
The early views of a simple, continuous stratigraphy have been revised by subsequent studies, that established (1) significant structural
repetitions in the belt; (2) the existence of distinct tectono-stratigraphic domains in faulted
contact (De Wit et al., 1983, 1992; Armstrong
et al., 1990; De Ronde and De Wit, 1994;
De Ronde and Kamo, 2000; De Wit, 1983;
3. The Moodies (ca. 3.23—3.21 Ga) group Kamo and Davis, 1994; Kröner et al., 1996;
consists of mostly coarse-grained clastic Lamb, 1984; Lowe, 1999, 1994; Lowe and Bysediments (Lowe and Byerly, 1999).
erly, 1999).
2. The Fig Tree group (ca. 3.26—3.22 Ga)
is composed of fine-grained, detrital sediments, with minor chemical sediments
and volcanoclastic rocks (Lowe and Byerly, 1999).
1. Stratigraphy of the Barberton Greenstone Belt
5
Figure 2: Simplified log in the Barberton Greenstone Belt (Lowe and Byerly, 1999). Different domains
show different stratigraphies; the most important break occurs between the South-(Eastern) and the North(Western) domain.
6
2. Main tectono-magmatic events
The BGGT formed between ca. 3.51 and 3.11
Ga1 and although supracrustal rocks (lavas and
sediments) from the belt itself yield a relatively
continuous spread of ages from 3559 ± 27 Ma
(Byerly et al., 1996; Poujol et al., 2003) to
3164 ± 12 Ma (Armstrong et al., 1990; Poujol et al., 2003), the BGGT was predominantly assembled during three or four discrete
tectono-magmatic events at 3.55—3.49, 3.49—
3.42, 3.255—3.225 and 3.105—3.07 Ga (Poujol
et al., 2003).
The correlation of deformational events between different parts of the BGGT is still controversial, but there is general consensus that
the geology of the belt mostly reflects assembly during two main phases of deformation in
an accretionary environment that occurred between 3.49 and 42 Ga (D1 ), and 3.25—3.21 Ga
(D2 ) (Lowe, 1999)2 . The early history of the
belt (D1 and pre-D1 , 3.49—3.55 Ga) is well
represented in the Swaziland Ancient Gneiss
Complex to the East. However, in the BGGT
proper, > 3.42 Ga rocks are restricted to the
high-grade “Stolzburg domain”.
The formation of plutonic rocks is associated
with each of the proposed tectono-magmatic
events. An early, “pre-D1 ” event (3.55—3.49
Ga) is recorded by the Steynsdorp pluton emplaced at the southern tip of the BGB. The
ca. 3.45 Ga plutons from the Stolzburg domain (Stolzburg, Theespruit, and minor plutons further South) were emplaced during the
D1 event. The three large plutons that form the
north-western margin to the belt (Badplaas,
Nelshoogte and Kaap Valley) formed during
the D2 event. The formation of the ca. 3.11
Ga plutons of the GMS suite has been considered to have occurred within an anorogenic
context. However, it has recently been demonstrated (Belcher and Kisters, 2006,b; Sonke,
2006; Westraat et al., 2004) that the emplacement of these plutons is associated with conjugate shearing and bulk crustal shortening
(termed D3 ).
1 Ages indicated in millions of years (Ma) correspond to actual, measured ages with reference and error, while
dates given in billions of years (Ga) refer to generalized time intervals.
2 In Lowe’s (1999) terminology, 5 successive deformation events (D to D ) are reported. D is ca. 3.45 Ga,
1
5
1
D2 to D5 ) are ca. 3.2 Ga (Fig Tree and Moodies age). For clarity, we use here a simpler terminology: D1 (ca.
3.45 Ga); D2 (3.25—3.20 Ga), separated in D2a (Fig Tree time) and D2b (Moodies time); D3 (ca. 3.1 Ga),
emplacement of the GMS suite
3. Accretion stage of the BGGT at > 3.42 Ga (pre-D1 and D1 )
7
3. Accretion stage of the BGGT at > 3.42 Ga (pre-D1 and D1 )
The 3.42—3.49 Ga event corresponds to
the formation of the Komati, Hooggenoeg
and Kromberg Formations of the Onverwacht
Group (Lowe, 1999; Lowe and Byerly, in press,
1999, and references therein). These three formations predominantly consist of mafic to ultramafic lavas, with subordinate cherts.
b. Associated granitoids
The ca. 3.55—3.50 Ga Steynsdorp pluton
(Kröner et al., 1996; Robb and Anhaeusser,
1983) consists of banded gneisses with a pervasive solid-state gneissosity that outcrop in a domal antiform (Kisters and Anhaeusser, 1995b)
and is in tectonic contact with the enveloping
supracrustal sequence of the Theespruit formation. The protolith of the Steynsdorp gneisses
is tonalitic (Kisters and Anhaeusser, 1995b;
Kröner et al., 1996), although a granodioritic
component, possibly related to the re-melting
Figure 3: 3.55—3.35 Ga history of the BGB (Lowe, of older tonalites, is also recorded.
1999): an oceanic plateau modified by subsequent
subduction and rifting.
3.2. . . . modified by a subduction (?)
event
3.1. An oceanic plateau. . .
a. Geology
The > 3.5 Ga event is represented by the mafic
and felsic volcanics of the Theespruit formation (Lowe and Byerly, in press, 1999, and references therein), which are coeval with the emplacement of the ca. 3.55—3.50 Ga Steynsdorp
pluton (Kröner et al., 1996). Little information
is available regarding the geological context of
their formation. However, the abundance of
mafic-ultramafic magmatism, together with the
geochemistry of the Steynsdorp pluton, suggesting shallow melting of composite (crustal
?) sources (Moyen et al., in prress), is consistent with an intraplate (oceanic plateau?) context, or bimodal magmatism associated with
rifting.
a. The D1 event
At the contact between the Hooggenoeg and
Kromberg Formations, the ca. 3.44-3.45 Ga
“H6” unit (Armstrong et al., 1990; Byerly et al.,
1996; Kröner et al., 1991; Kröner and Todt,
1988) is nearly synchronous with the intrusion
of the TTG plutons from the Stolzburg domain (Theespruit, Stolzburg, and the minor
plutons to the South defined by Anhaeusser
et al., 1981). The H6 unit is a thin (few tens
of meters), unit of dacitic lava flows and shallow intrusive bodies (geochemically regarded as
the extrusive equivalents of the TTG plutons,
De Wit et al., 1987), as well as clastic sediments
and conglomerates. This suggests that some topography existed at this stage in the evolution
of the belt. The first, well constrained deformation event affecting the belt (D1 Lowe et al.,
1999) also occurred at about the same time.
8
The geochemistry of the ca. 3.45 Ga TTG plutons suggests deep melting (15–20 kbar) of depleted (oceanic crust?) sources, at relatively
cold temperatures (900 ◦ C), which is consistent
with a subduction scenario. Collectively, these
lines of geological evidence suggest the development of a transient active margin (oceanic arc)
(De Ronde and Kamo, 2000; Lowe, 1999; Lowe
and Byerly, in press, , and references therein)
at ca. 3.45 Ga, fringing the previously assembled oceanic plateau. This would result in magmatic modifications of the plateau, eventually
leading to the stabilization of a continental nucleus (Benn and Moyen, in press; White et al.,
1999). This early continental nucleus may then
behave in a rigid, coherent way during the subsequent events.
b. Associated granitoids
The ca. 3.45 Ga (syn-D1 ) TTGs are represented by a number of intrusive bodies in
the Stolzburg terrane located to the south of
the greenstone belt (Anhaeusser and Robb,
1980; Kisters et al., 2003; Moyen et al.,
2006; Robb and Anhaeusser, 1983; Viljoen
and Viljoen, 1969a). The two most prominent and better defined intrusions are the
Stolzburg and Theespruit plutons that intruded
the supracrustal rocks of the belt. Further
south, several smaller plutons or domains are
recognized and form a complex pattern of TTG
gneisses and greenstone remnants, partially
transposed and dismembered by ca. 3.1 Ga
shear zones. These are the Theeboom, Eerstehoek, Honingklip, Weergevonden “cells” and
“plutons” of Anhaeusser et al. (1981); Robb
and Anhaeusser (1983).
The plutons preserve clearly intrusive relations
with the surrounding greenstones (Kisters and
Anhaeusser, 1995a; Kisters et al., 2003, figure
4), although the terrane as a whole (granitoids and country rocks) were deformed during the D2 orogeny (Diener et al., 2006, 2005;
Kisters et al., 2003; Stevens and Moyen, in
press). The nature of the preserved contacts,
occasionally occurring as intrusive breccias, the
presence of a network of surrounding dykes,
the existence of simultaneous, cogenetic extrusive rocks, all suggest that the Stolzburg pluton
(and the other plutons of the terrane/domain)
intruded at shallow levels under brittle conditions (Kisters and Anhaeusser, 1995a). All
the ca. 3.45 Ga plutons are composed predominantly of medium- and/or coarse-grained
leucotrondhjemites (Kisters and Anhaeusser,
1995a; Robb and Anhaeusser, 1983; Yearron,
2003; Moyen et al., in prress).
Figure 4: Intrusive relations of the Theespruit pluton with surrounding greenstones (Kisters and Anhaeusser,
1995a).
4. A subduction-collision-exhumation orogen at 3.29-3.21 Ga (D2 )
9
3.3. Renewed (ultra)mafic activity
the Western terranes (the Eastern and Western terranes and their significance is discussed
Following the D1 event, the Mendon Forma- below), from ca. 3.42 to 3.25 Ga. Based on
tion was deposited in the Stolzburg domain studies of the volcanic and sedimentary units, a
(Songimvelo and Steynsdorp blocks) in the east period of quiescence (rift/intracontinental set(Lowe, 1999), and the Weltvreden Formation in ting) is suggested (Lowe, 1999).
The dominant geological event that shaped
the present-day structure of the belt occurred
during the D2 events3 , at ca. 3.25—3.21
Ga. Structural (De Ronde and De Wit, 1994;
De Ronde and Kamo, 2000; De Wit et al.,
1992; Kisters et al., 2003) as well as metamorphic (Diener et al., 2006, 2005; Dziggel et al.,
2005, 2002; Kisters et al., 2003; Moyen et al.,
2006; Stevens et al., 2002) studies suggest the
collision (or arc accretion) between two relatively rigid blocks, separated by the Inyoni–
Inyoka tectonic system (Lowe, 1994). The western terrane has largely been overprinted by the
ca. 3.25–3.21 Ga rocks (Fig Tree lavas and
TTGs), but was probably formed on a nucleus
of slightly older (3.3–3.25 Ga; De Ronde and
De Wit, 1994; Lowe, 1994; Lowe and Byerly,
1999; Lowe et al., 1999; De Ronde and Kamo,
2000) mafic and ultramafic lavas. The eastern terrane is better preserved; it was at this
time a composite continental nucleus including
old lavas and sediments, intruded by the ca.
3.45 Ga TTGs, and overlain by still younger
mafic/ultramafic lavas: an oceanic plateau,
modified by a relatively minor subduction event
(see above). The accretion itself occurred via
under-thrusting (subduction? and the eastern,
high-grade Stolzburg terrane probably represented the lower plate of this event.
of the BGB (Anhaeusser et al., 1981, 1983;
De Ronde and De Wit, 1994; De Ronde and
Kamo, 2000; De Wit et al., 1992; Lowe et al.,
1999; Lowe, 1994; Lowe and Byerly, 1999;
Viljoen and Viljoen, 1969b). To the West, the
Onverwacht group is mostly 3.3—3.25 Ga old,
whereas it is much older in the East (3.55—
3.3 Ga). Furthermore, the details of the stratigraphic sequences on both sides cannot be correlated, indicating that the two parts of the belt
evolved independently. The boundary between
the two domains is tectonic and corresponds to
the Inyoka–Saddleback fault system described
below. This structure spans the length of the
belt from the “Stolzburg arm” near Badplaas
in the South, to the Northern extremity (Kaapmuiden).
b. Tectonic history of the BGB
Four (or five) successive deformation phases related to the 3.25—3.20 Ga D2 period are identified. The first occurred during the deposition
of the sediments and felsic volcanics of the Fig
Tree group, at 3.25–3.23 Ga, probably associated with the development of a volcanic arc in
what is now the terrane West of the InyoniInyoka fault system. At ca. 3.23 Ga, a subsequent dominant period of deformation resulted
from the accretion of the two terranes along the
4.1. Evidence for accretionary orogen Inyoni-Inyoka fault system.
in the BGGT
a. Stratigraphy
The Onverwacht (and, to some degree, the Fig
Tree) groups show different stratigraphies in
the North-Western, and South-Eastern parts
3 Encompassing
the D2 to D5 events of Lowe (1999)
10
Figure 5: The different terranes composing the BGB
(Lowe, 1994). The main break is the Inyoka fault
A consequence of the D2 accretion was, at ca.
3.22—3.21 Ga, the syn-tectonic deposition of
the sandstone and conglomerates of the Moodies group in small and discontinuous faultbounded basins (Heubeck and Lowe, 1994a,b).
These rocks were deposited, at least in part,
in extensional basins formed by normal faulting in the BGB (upper crust) in response to
core complex exhumation and diapiric rise of
gneissic domes in the lower crust (surrounding
granitoids) (Kisters et al., 2003, 2004). Thus,
they represent the sedimentological response to
post-collisional collapse. Finally, late ongoing
compression resulted is interpreted to produce
strike-slip faulting and folding of the whole sequence (including the Moodies group); the timing of this “late” compression is not known.
Figure 6: Synthetic logs in the different terranes (Lowe,
1994).
It corresponds to the limit between the Northwestern, and Southeastern facies of the Onverwacht group. Along the Inyoka fault, several
layered mafic/ultramafic complexes are found
(Anhaeusser, 2001), that could correspond to
fragments of oceanic crust trapped in a suture zone. On a larger scale, this zone corresponds to a geophysical boundary within the
Kaapvaal craton, that runs for several hundreds of kilometers and separates two geophysicaly and geochronologicaly distinct terranes
(De Wit et al., 1992; Poujol, in press; Poujol
et al., 2003). The Inyoka fault zone itself is
made of a network of subvertical faults, which
were active during several of the later deformation events described above, leading to a complex history. It is interpreted as a D2 thrust,
that was steepened during subsequent deformation.
c. The Inyoka-Inyoni fault system
Within the BGB, the main D2 structure corresponds to a fault zone parallel to the belt edge,
in its Northwestern part. This tectonic structure is called the “Inyoka-Saddleback fault”
(Lowe, 1999; Lowe et al., 1999; Lowe, 1994).
Further South, external to the main BGB and
in the granitoid dominated terrane , a ductile,
North-South trending shear zone runs from the
extremity of the Stolzburg “arm” of the BGB,
towards the Schapenburg Schist belt some 30
km further South. This zone, called the “In-
yoni shear zone” (ISZ; Kisters et al., 2004;
Moyen et al., 2006; Stevens et al., 2002), is a
major structure in the granitoid terrane South
of the BGB; it separates the ca. 3.2 Ga Badplaas gneisses to the West, from the ca. 3.45
Ga Stolzburg pluton in the East, mirroring the
11
difference between the relatively young, western “Kaap Valley” block and the older terranes
(Songimvelo, etc., Lowe, 1994) to the East
of the Inyoka fault. Thus, the ISZ is probably a lower crustal equivalent of the Inyoka–
Saddleback fault system.
Figure 7: Delineation of tectonic terranes in the BGB (Stevens and Moyen, in press). Boxes refer to the
site of detailed metamorphic studies (see page 78 in appendix). Suture zone: ISZ= Inyoni Shear Zone, IF=
Inyoka fault. Detachments: KaF= Kaap River fault, Ko=komati fault.
12
4.2. Metamorphism associated with
the ca. 3.2 Ga orogen
Recent metamorphic studies in the BGGT gave
results summarized in a paper reproduced in
appendix (article page 78). The main conclusions of these studies can be summarized as follows:
• The relatively low-grade greenstone belt
is separated from higher grade basement
(lower to middle crust) by a sharp metamorphic break with a pressure transition
of > 5 kbar (ca. 15 km) over just a few
kilometers laterally. This transition occurs in zones of high-strain rocks (up to
mylonites), that record a normal sense of
movement with the low-grade greenstone
belt being down-thrown relative to the
surrounding amphibolite-facies gneisses.
In essence, these zones define the cuspate granite-greenstone contacts of the
“dome and keel” pattern. Peak metamorphism in these areas is syntectonic with
the exhumation process, which is continuous into the greenschist facies.
• In the high-grade domains away from
the contact with the lower grade belt,
peak metamorphic conditions are posttectonic.
• Two different thermal regimes are
recorded in the deep crust of the BGGT.
Mid- to lower-crustal rocks from the
North-Western domain generally record
metamorphic field gradients as low as 18
to 20 ◦ C.km−1 . Similar rocks from the
South-Eastern domain record metamorphic field gradients of 30 to 40 ◦ C.km−1 .
Such a duality of metamorphic regimes
has been described as a “hallmark of plate
tectonics” (Brown, 2006); in this sense to
appears that the BGGT represents a dismembered (principally by exhumation of
the high grade crust) system of paired
metamorphic belts developed at ca 3.23
Ga.
• The most compelling metamorphic evidence for an accretionary orogen comes
from the Inyoni shear zone, the lower
crustal expression of the main terrane
boundary described in the belt. The
pressures reported for this zone are, at
present, the highest crustal pressures reported for Archaean rocks, and correspond to by far the lowest known apparent geothermal gradients (12 ◦ C.km−1 ) in
the Archaean rock record. In the modern Earth, the only process capable of
producing crustal rock evolution through
this P–T domain occurs within subduction zones
4.3. Ca. 3.2 Ga magmatism
The 3.29-3.21 Ga syn D2 TTG plutons form
a composite group that occurs along the northern and southwestern margins of the Barberton
Belt (Anhaeusser and Robb, 1980; Robb and
Anhaeusser, 1983; Viljoen and Viljoen, 1969a;
Moyen et al., in prress).
• In the south, the 3290—3240 Ma (Kisters
et al., 2006, Kisters et al., in prep) Badplaas gneisses (and probably the apparently similar Rooihoogte gneisses, west
of the 3.1 Ga Heerenveen batholith) consist of two main components. These
include an older, coarse grained leucotrondhejmitic rock that underwent solidstate defomation and a younger, multiphase intrusive component, made up a
variety of typically finer grained trondhjemites. In proximity to the Inyoni shear
zone, the main suture in the southern
TTG gneiss terrain, most of these intrusions are syntectonic. Further away from
the shear zone, the trondhjemites form
either irregularly shaped, discontinuous,
stockwork-like breccias or small (100 m
– 5 km) plugs and intrusions. The longlived emplacement of the Badplaas pluton, and its composite nature, makes it
unique in the BGGT.
• Further north, the composite 3.23—3.21
Ga Nelshoogte pluton (Anhaeusser et al.,
1981, 1983; Belcher et al., 2005; Robb
and Anhaeusser, 1983) is dominated by
coarse-grained leuco-trondhjemites, that
are intruded by amphibole-tonalites, particularly along the northern and northeastern margin of the pluton. The pluton was intruded during regional folding,
probably as a laccolith, and lit-par-lit intrusive relations , as well as smaller-scale
brecciation with the surrounding greenstone wallrocks are preserved (Belcher
et al., 2005). This is again suggestive
of relatively shallow emplacement of the
Nelshoogte pluton. The domal map pattern around the Nelshoogte pluton reflects late stage folding and steepening
of the syn-emplacement, initially flat fabrics.
• The large 3.23—3.22 Ga Kaap Valley pluton along the northern margin of the Barberton Belt is, for the most part, made
up of coarse-grained, biotite-amphibole
tonalite (Robb et al., 1986), with minor
occurrences of amphibole-tonalite (biotite free).
In the context of the ca 3.2 Ga orogenic history,
the ca. 3.29—3.21 Ga plutons record the transition from pre-collision to post-collision mag-
13
matism. The earliest phases formed by deep
melting (trondhjemites forming parts of the
Badplaas gneisses, and most of the Nelshoogte
pluton), and their ages corresponds to the accretion stage of the BGGT, most probably in
a magmatic arc (Martin, 1986; Kisters et al.,
2006). The latest phases (Parts of Badplaas
gneisses, tonalitic components in the Nelhoogte
pluton, Kaap Valley pluton) formed by relatively shallow (10-12 kbar) melting of amphibolites, possibly parts of the Onverwacht Group.
The transition from trondhjemites (bulk of the
Nelshoogte pluton, part of Badplaas gneisses)
to tonalites (late phases in the Nelshoogte pluton, Kaap Valley pluton) reflects increasing
temperatures at the base of the collapsing pile,
as commonly observed in post-orogenic collapse (Kisters et al., 2003). Some of the early
formed rocks of the Badplaas pluton underwent intracrustal remelting shortly after their
emplacement. Field and structural studies
demonstrate that at least some of these plutons
formed during orogen parallel extension. Collectively this evidence is consistent with lower
crustal melting of the thickened, dominantly
mafic crust during orogenic collapse, and/or
possibly during slab breakoff.
Figure 8: Emplacement model for the Nelshoogte pluton (Belcher et al., 2005).
14
4.4. Evolution model
The model proposed for the ca. 3.2 Ga orogenic
history is the following:
• From ca. 3.25 to 3.23 Ga, syn-tectonic
deposition of the felsic volcanics and clastic sediments of the Fig Tree Group,
probably resulting in the development of
a volcanic arc in what is now the terrane west of the Inyoni-Inyoka fault system (De Ronde and Kamo, 2000; Kisters
et al., 2006; Lowe, 1999). The Badplaas
gneisses were emplaced in the western terrane during this period.
• At ca. 3.23 Ga, the main tectonic phase
results in the accretion of the two terranes
along the Inyoni-Inyoka fault system.
This is accompanied by high-pressure,
low to medium-temperature metamorphism of the eastern, Stolzburg domain
(Diener et al., 2005; Dziggel et al., 2002;
Moyen et al., 2006), especially along the
fault system, interpreted as a suture zone
(Stevens and Moyen, in press).
• The collision is immediately followed at
ca. 3.22—3.21 Ga, by the extensional collapse of the orogenic pile (Kisters et al.,
2003), leading to the nearly isothermal
exhumation of the high-pressure rocks of
the Stolzburg domain along detachment
faults (Diener et al., 2005; Moyen et al.,
2006) and the emplacement of a new set
of TTG plutonic rocks (Nelshoogte and
Kaap Valley plutons). The extensional
collapse is synchronous with the deposition of (at least part of) the detrical
Moodies Group in small, discontinuous,
maybe fault-bounded basins (Heubeck
and Lowe, 1994a,b). This is immediately
followed by diapiric exhumation of the
lower crust, and steepening of the fabrics.
• Finally, late ongoing deformation resulted
in strike-slip faulting and folding of the
whole sequence (including the Moodies
Group). Some late to post-tectonic plutons (e.g. Dalmein, 3215 ± 2 Ma; Kamo
and Davis, 1994), crosscutting all ca.
3.23—3.21 Ga structures, also form during this period.
15
Figure 9: Geodynamic model of evolution for the 3.29—3.21 Ga accretionary orogen. Points A, B and C
(and their respective P–T path, on the right hand side) correspond to respectively the Tjakastad Schist Belt
(stop 3.2), the Inyoni Shear Zone (stop 3.4) and the Schapenburg Schist Belt (stop 5.4)
16
5. The sheeted batholiths of the GMS suite (3.11 Ga)
374
J. D. W E S T R A AT E T A L .
Fig. 1. Regional geology of the
granite–greenstone terrane (afte
Anhaeusser et al. 1981) and its
the Kaapvaal Craton in southern
(inset).
Figure 10: Extend of the ca. 3.1 Ga GMS suite (Westraat et al., 2004).
anorogenic emplacement of the granitoids (e.g. Anh
sequences (e.g. De Wit et al. 1987; Armstrong et al. 1990); or
Robb 1983). This interpretation has not remained un
(3) questioning the magmatic models for large parts of the
and Robb et al. (1983) and Jackson & Roberts
present-day granite–greenstone contacts altogether, as structudescribed
presence
regional-scale
gneiss belts
rally reworkeddominant
and subsequently
exhumed
The volumetrically
intrusions
in basement
similar gneisses
to the 3.5—3.2
Ga the
TTG
rocksof of
the
along
margins1983;
of theRobb,
batholiths. The multiphas
(e.g.belong
Dziggel ettoal.the
2002;“GMS”
Kisters et(graniteal. 2003). BGGT (Anhaeusser
the BGGT
andtheRobb,
relationships between basement gneisses and the GMS
This study focuses on laterally extensive granite plutons of a
monzonite-syenite)
suite (Yearron, 2003), and 1983) (2) a syenitic
to syenogranitic compodeformation of the potassic granitoids suggests that th
subsequent magmatic episode associated with the intrusion of
were emplaced
at
ca.
3.11
Ga
(Kamo
nent,
evident
in
the
Boesmanskop
(Anhaeusser
ment of the 3.1 Ga
granitoids is, at least partly,
vast amounts of granodiorites, monzogranites and syenites, the
and Davis,
1994;
Maphalala
and
Kröner,
et
al.,
1983)
and
Kees
Zyn
Doorn
intrusions,
controlled.
As
a
result
of these contrasting views on
GMS suite, at c. 3.1 Ga. Rocks of the GMS suite are found not
relationships
and the lack
of detailed
onlyare
in the
Barberton granite–greenstone
terrane,
alsopresent
over e.g.
1993). They
represented
by the Piggs’Peak
butbut
also
on the Western
margin
of structural work o
batholiths,
the emplacement
and tectonic setting of
the Kaapvaal
Craton, and their
batholith large
(east parts
of theofBGGT
and in Swaziland),
theemplacement
Mpuluzi batholith
(Westraat
et al., 2004).
plutonic
suite have remains
remained somewhat enigmatic
with(in
the the
first stabilization
of the
of the
Nelspruit coincides
batholith
north), and
thecentral
Theparts
origin
of the wide
second
component
The present study centres around an area of c. 40 k
craton (De Wit et al. 1992; Kamo & Davis 1994; Poujol &
Mpuluzi/Lochiel
and Heerenveen batholiths (in poorly constrained.along the western and northern margin of the Mesoa
Anhaeusser 2001). The GMS suite in the Barberton granite–
the south).
Collectively,
they
mostly
3105 Ma Mpuluzi batholith, one of the most extensiv
greenstone terrane shows
veryare
different
internal and external
The potassic
the GMS
suite
leucogranites,
granitesfrom
and the
granodiorites,
plutons from
of the GMS
suite (Anhaeusser
et al. 1981; An
characteristics
earlier TTG assosuite. Individual
plutons batholiths
form
laterally
extensive,
flat
intrusions
(probRobb 1983; Kamo & Davis 1994; Yearron 2003) (Fig
mayminor
cover monzonites
several thousand
square
kilometres and these compociated with
and
syenites.
no tomore
kmthisthick).
The1–2
aim of
study is toTheir
constrain the emplacemen
site granitoid bodies have traditionally been ably
referred
as than
isms
and partially
magmatic assembly
of this large batholith
batholiths,
alluding
to their
texturally het- was at
Rocks from
the GMS
suite
are compositionally
distinctly bi- andemplacement
least
guided by
bines a number
of internalshear
and external structural fe
erogeneous nature
enormous
extent (e.g.
et conjugate
modal (Anhaeusser
and and
Robb,
1983;areal
Belcher
a Anhaeusser
network of
syn-magmatic
seem typical of many of the GMS suite plutons (R
al. 1981). For the most part, the plutons appear undeformed,
2006,b; Westraat et al., 2004; Year- zones, reflecting an1983).
and Kisters,
event This
of craton
wide commargin, in particular, discloses high
intrusion-related wall-rock strains are only locally recorded, and
ron, 2003),
with
(1)
a
leucogranitic
to
granpression.
One
such
shear
zone
(the
Welvercontact relationships between
the younger GMS suite
intrusive relationships with wall rocks are commonly sharply
odioritic component,
probably
originating
from
Westraat
et
al.,
2004)
hasreflect the existing c
diend
shear
zone,
basement gneisses that closely
discordant (e.g. Hunter 1973; Anhaeusser & Robb 1983; Robb et
about
syn- v. post-tectonic
timing and controls of
al. 1983).
Regional
studiescompositionally
have demonstrated that
mostdocumented
of these
the partial
melting
of rocks
been
on
thethemargin
of the Mpuemplacement (Anhaeusser & Robb 1983; Jackson &
granitoids represent subhorizontal, sheet-like intrusions. The
1983). Mapping was undertaken on the basis of ae
tabular granites are commonly underlain by so-called migmatite
graphs at a scale of between 1:6000 and 1:10 000, a
terranes and dyke complexes that have tentatively been interand spatial distortions were corrected by global
preted as the feeders to the overlying granite sheets (e.g. Hunter
system (GPS) readings. The field-based studies were s
1957, 1973; Anhaeusser et al. 1981; Anhaeusser & Robb 1983;
ted by thin-section petrography and whole-rock geoch
Robb et al. 1983). The sum of these features has traditionally
characterize different intrusive phases. In addition, ge
been interpreted to indicate a ‘passive’, post-tectonic and
Bibliography
17
luzi batholith; it merges with the southern ex- structures. The exact extent of the ca. 3.1
tremity of the Inyoni Shear zone, that is there Ga structural overprint on the older rocks and
rotated into parallelism with the ca 3.11 Ga structures remains poorly understood.
Bibliography
Anhaeusser, C. and Robb, L. (1980). Regional and detailed field and geochemical studies of
archean trondhjemitic gneisses, migmatites and greenstone xenoliths in the southern part of
the Barberton mountain land, South Africa. Precambrian Research, 11:373–397.
Anhaeusser, C., Robb, L., and Viljoen, M. (1981). Provisonnal geological map of the Barberton
greenstone belt and surrounding granitic terrane, eastern Transvaal and Swaziland (1:250 000).
Anhaeusser, C., Robb, L., and Viljoen, M. (1983a). Notes on the provisionnal geological map
of the Barberton greenstone belt and surrounding granitic terrane, eastern Transvaal and
Swaziland (1:250 000 color map). Transactions of the Geological Society of South Africa,
9:221–223.
Anhaeusser, C. R. (1969). The stratigraphy, structure and gold mineralization of the Jamestown
abd Sheba hills area of the Barberton Mountain Land. PhD thesis, University of the Witswatersrand, Johannesburg.
Anhaeusser, C. R. (2001). The anatomy of an extrusive-intrusive Archaean mafic-ultramafic
sequence: the Nelshoogte Schist Belt and Sttolzburg layered ultramafic complex, Barberton
Greenstone Belt, South Africa. South African Journal of Geology, 104:167–204.
Anhaeusser, C. R. and Robb, L. J. (1983). Chemical analyses of granitoid rocks from the
Barbeton Mountain Land. Geological Society of South Africa Special Publication, 9:189–219.
Anhaeusser, C. R., Robb, L. J., and Barton, J.M., J. (1983b). Mineralogy, petrology and origin f
the Boesmanskop syeno-granite complex, Barberton mountain land, South Africa. Geological
Society of South Africa Special Publication, 9:169–183.
Armstrong, R., Compston, W., De Wit, M. J., and Williams, I. (1990). The stratigraphy of the
3.5-3.2 Ga Barberton greenstone belt revisited: a single zircon ion microprobe study. Earth
and Planetary Science Letters, 101:90–106.
Belcher, R. W. and Kisters, A. F. (2006a). Progressive adjustments of ascent and emplacement
controls during incremental construction of the 3.1 Ga Heerenveen batholith, South Africa.
Journal of Structural Geology, 28:1406–1421.
Belcher, R. W. and Kisters, A. F. (2006b). Syntectonic emplacement and deformation of the
Heerenveen batholith: conjectures on the structural setting of the 3.1 Ga granite magmatism in
the Barberton granite-greenstone terrain, South Africa. Geological society of America special
Publication, 405:211–231.
Belcher, R. W., Kisters, A. F., Poujol, M., and Stevens, G. (2005). Structural emplacement of
the 3.2 ga Nelshoogte pluton: implications for the origin of dome-and-keel structures in the
Barberton granite-greenstone terrain. In Geocongress, Durban.
Benn, K. and Moyen, J.-F. (in press). Geodynamic origin and tectonomagmatic evolution of
the Late Archean Abitibi-Opatica terrane, Superior Province: Magmatic modification of a
plateau-type crust and plume-subduction interaction . In Condie, K., editor, Penrose special
volume GSA. GSA, Boulder.
18
Brown, M. (2006). A duality of metamorphic styles is the hallmark of plate tectonics. Geology,
34(11):961–964.
Byerly, G. R., Kroner, A., Lowe, D. R., Todt, W., and Walsh, M. M. (1996). Prolonged magmatism and time constraints for sediment deposition in the early Archean Barberton greenstone
belt: Evidence from the Upper Onverwacht and Fig Tree groups. Precambrian Research,
78(1-3):125–138.
De Ronde, C. and De Wit, M. (1994). The tectonic history of the Barberton greenstone belt,
South Africa: 450 million years of Archean crustal evolution. Tectonics, 13:983–1005.
De Ronde, C. and Kamo, S. (2000). An Archaean arc-arc collisional event: a short-lived (ca 3
Myr) episode, Weltvreden area, Barberton greenstone belt, South Africa. Journal of African
Earth Sciences, 30(2):219–248.
De Wit, M., Roering, C., Hart, R., Armstrong, R., De Ronde, C., Green, R., Tredoux, M.,
Peberdy, E., and Hart, R. (1992). Formation of an Archaean continent. Nature, 357:553–562.
De Wit, M. J. (1983). Notes on a preliminary 1:25,000 geological map of the southern part of the
Barberton greenstone belt. In Anhaeusser, C. R., editor, Contributions to the geology of the
Barberton Mountain land, volume 9 of Special publications, pages 185–187. Geological Society
of South Africa.
De Wit, M. J., Armstrong, R., Hart, R., and Wilson, A. (1987). Felsic igneous rocks within the
3.3 to 3.5 Ga Barberton Greenstone belt: high crustal level equivalents of the surrounding
tonalite-trondhjemite terrain, emplaced during crustal thrusting. Tectonics, 6:529–549.
De Wit, M. J., Fripp, R., and Stanistreet, I. (1983). Tectonic and stratigraphic implications of
new field observations along the southern part of the Barberton greenstone belt. In Anhaeusser,
C. R., editor, Contributions to the Geology of the Barberton Mountain Land, volume 9 of Special
publications, pages 21–29. Geological Society of South Africa.
Diener, J., Stevens, G., and Kisters, A. F. (2006). High-pressure low-temperature metamorphism
in the southern Barberton granitoid greenstone terrain, South Africa: a record of overthickening and collapse of Mid-Archaean continental crust. In Benn, K., Mareschal, J.-C., and Condie,
K., editors, Archean Geodynamic Processes, volume 164 of monographs, pages 239–254. AGU.
Diener, J., Stevens, G., Kisters, A. F., and Poujol, M. (2005). Metamorphism and exhumation
of the basal parts of the Barberton greenstone belt, South Africa: Constraining the rates of
mid-Archaean tectonism. Precambrian Research, 143:87–112.
Dziggel, A., Armstrong, R., Stevens, G., and Nasdala, L. (2005). Growth of zircon and titanite
during metamorphism in the granitoid-gneiss terrain south of the Barberton greenstone belt,
South Africa. Mineralogical Magazine, 69:1021–1038.
Dziggel, A., Stevens, G., Poujol, M., Anhaeusser, C., and Armstrong, R. (2002). Metamorphism
of the granite-greenstone terrane South of the Barberton greenstone belt, South Africa: an
insight into the tectono-thermal evolution of the ’lower’ portions of the Onverwacht group.
Precambrian Research, 114:221–247.
Heubeck, C. and Lowe, D. R. (1994a). Depositional and Tectonic Setting of the Archean Moodies
Group, Barberton Greenstone-Belt, South-Africa. Precambrian Research, 68(3-4):257–290.
Heubeck, C. and Lowe, D. R. (1994b). Late Syndepositional Deformation and Detachment
Tectonics in the Barberton Greenstone-Belt, South-Africa. Tectonics, 13(6):1514–1536.
Kamo, S. L. and Davis, D. W. (1994). Reassessment of Archean Crustal Development in the
Barberton Mountain Land, South-Africa, Based on U-Pb Dating. Tectonics, 13(1):167–192.
Bibliography
19
Kisters, A. F. and Anhaeusser, C. (1995a). Emplacement features of Archaean TTG plutons along
the Southern margin of the Barberton greenstone belt, South Africa. Precambrian Research,
75:1–15.
Kisters, A. F. and Anhaeusser, C. R. (1995b). The structural significance of the Steynsdorp
pluton and anticline within the tectono-magmatic framework of the Barberton Mountain land.
South African Journal of Geology, 98:43–51.
Kisters, A. F., Belcher, R. W., Poujol, M., Stevens, G., and Moyen, J.-F. (2006). A 3.2 Ga
Magmatic arc Preserving 50 Ma of Crustal Convergence in the Barberton Terrain, South
Africa. In AGU fall meeting, pages V11D–0619.
Kisters, A. F., Stevens, G., Dziggel, A., and Armstrong, R. (2003). Extensional detachment
faulting and core-complex formation in the southern Barberton granite–greenstone terrain,
South Africa: evidence for a 3.2 Ga orogenic collapse. Precambrian Research, 127:355–378.
Kisters, A. F., Stevens, G., and Van Reenen, D. (2004). Excursion field guide: the Kaapvaal
traverse. 17-23 july 2004. In Geoscience Africa, University of Witswaterrand, Johannesbourg.
Kröner, A., Byerly, G., and Lowe, D. (1991). Chronology of early Archaean granite-greenstone
evolution in the Barberton mountain land, South Africa, based on precise dating by single
zircon evaporation. Earth and Planetary Science Letters, 103:41–54.
Kröner, A., Hegner, E., Wendt, J., and Byerly, G. (1996). The oldest part of the Barberton
granitoid-greenstone terrain, South Africa: evidence for crust formation between 3.5 and 3.7
Ga. Precambrian Research, 78:105–124.
Kröner, A. and Todt, W. (1988). Single zircon dating constraining the maximum age of the
Barberton Greenstone Belt, South Africa. Journal of Geophysical Research, 93:15329–15337.
Lamb, S. (1984). Geology of part of the Archaean Barberton Greenstone Belt. PhD thesis,
Cambridge University.
Lowe, D. (1994). Accretionary history of the Archean Barberton Greenstone Belt (3.55-3.22 Ga),
southern Africa. Geology, 22:1099–1102.
Lowe, D. and Byerly, G. (1999). Stratigraphy of the west-central part of the Barberton greenstone
belt, South Africa. Geological Society of America Special Paper, 329:1–36.
Lowe, D. R. (1999). Geological evolution of the Barberton greenstone belt and vicinity. Geological
Society of America Special Paper, 329:287–312.
Lowe, D. R. and Byerly, G. (in press). An overview of the geology of the Barberton greenstone belt
and vicinity: implications for early crustal development. In Van Kranendonk, M., Smithies,
R. H., and Bennet, V., editors, Earth’s oldest rocks, Developments in Precambrian Geology.
Elsevier.
Lowe, D. R., Byerly, G., and Heubeck, C. (1999). Structural divisions and development of the
west-central part of the Barberton Greenstone Belt. Geological Society of America Special
Paper, 329:37–82.
Maphalala, R. and Kröner, A. (1993). Pb-Pb single zircon ages for the younger Archaean granitoids of Swaziland. In Extd. Abstracts 16th Inernational Colloquium on African Geology, pages
201–206, Mababane, Swaziland.
Martin, H. (1986). Effect of steeper Archean geothermal gradient on geochemistry of subductionzone magmas. Geology, 14(9):753–756. Sep.
Moyen, J.-F., Stevens, G., and Kisters, A. F. (2006). Record of mid-Archaean subduction from
metamorphism in the Barberton terrain, South Africa. Nature, 443:559–562.
20
Moyen, J.-F., Stevens, G., Kisters, A. F., and Belcher, R. W. (in press). TTG plutons of
the Barberton granitoid-greenstone terrain, South Africa. In Van Kranendonk, M., Smithies,
R. H., and Bennet, V., editors, Earth’s Oldest rocks, Developments in Precambrian geology,
page Chapter X. Elsevier.
Poujol, M. (in press). An overview of the pre-Mesoarchaean rocks of the Kaapvaal Craton, South
Africa. In Van Kranendonk, M., Smithies, R. H., and Bennet, V., editors, The early rocks.
Elsevier.
Poujol, M., Robb, L., Anhaeusser, C., and Gericke, B. (2003). A review of the geochronological
constraints on the evolution of the Kaapvaal craton, South Africa. Precambrian Research,
127:181–213.
Robb, L., Barton, J.M., J., Kable, E., and Wallace, R. (1986). Geology, geochemistry and isotopic
characteristics of the Archaean Kaap Valley pluton, Barberton mountain land, South Africa.
Robb, L. J. (1983). Geological and geochemical characteristics of late granite plutons in the Barberton region and Swaziland, with an emphasis on the Dalmein pluton — a review. Geological
Robb, L. J. and Anhaeusser, C. R. (1983). Chemical and petrogenetic characteristics of Archaean
tonalite-trondhjemite gneiss plutons in the Barberton mountain land. Geological Society of
South Africa Special Publication, 9:103–116.
Sonke, G.-J. (2006). Internal architecture of the sheeted margin of the 3.1 Ga Mpuluzi batholith,
Barberton granite-greenstone terrain. Honors thesis, Stellenbosch University.
Stevens, G., Droop, G., Armstrong, R., and Anhaeusser, C. (2002). Amphibolite-facies metamorphism in the Schapenburg schist belt: a record of the mid-crustal response to 3.23 Ga terrane
accretion in the Barberton greenstone belt. South African Journal of Geology, 105:271–284.
Stevens, G. and Moyen, J.-F. (in press). High-pressure, low-temperature metamorphism in the
Barberton greenstone belt; a key to understanding Archaean tectonic evolution. In Van Kranendonk, M., Smithies, R. H., and Bennet, V., editors, Earth’s oldest rocks, Developments in
Precambrian Geology. Elsevier.
Viljoen, M. and Viljoen, R. (1969a). A proposed new classification of the granitoid rocks of the
Barberton region. Geological Society of South Africa Special Publication, 2:153–188.
Viljoen, M. and Viljoen, R. (1969b). An introduction to the geology of the Barberton granitegreenstone terrain. Geological Society of South Africa Special Publication, 2:9–28.
Visser, D. (1956). The geology of the Barberton area, volume 15 of Special publication. Geological
Survey of South Africa.
Westraat, J. D., Kisters, A. F., Poujol, M., and Stevens, G. (2004). Transcurrent shearing, granite
sheeting and the incremental construction of the tabular 3.1 Ga Mpuluzi batholith, Barberton
granite-greenstone terrane, South Africa. Journal of the Geological Society of London, 161:1–
16.
White, R., Tarney, J., Kerr, A., Saunders, A., Kempton, P., Pringle, M., and Klaver, G. (1999).
Modification of an oceanic plateau, Aruba, Dutch Caribbean: implications for the generation
of continental crust. Lithos, 46:43–68.
Yearron, L. (2003). Archaean granite petrogenesis and implications for the evolution of the
Barberton mountain land, South Africa. Unpub. phd thesis, Kingston University.
Part II
Field itinerary
Day 1: Saturday, 7 July
Transfert from Stellenbosch to Johannesburg and Badplaas
Depart Capetown Airport (domestic terminal) on flight 1 Time IT 104 at 12h20. Arrive Johannesburg at 14h20.
Figure 11: Road map, from Johannesburg to Badplaas. The dotted box corresponds to the extend of the
route map, page iv
Drive out of Johannesburg airport on the R21
southbound (direction Boksburg). Turn left after 4 km on N12 (direction Witbank). We now
drive in Johannesburg Eastern suburbs, in the
East Rand goldfields — as evidenced by the
gold mine dumps, of a pale yellow color. The
area is underlain by the Witswatersrand Supergroup (3.0 – 2.7 Ga) sediments, commonly
covered by the Karoo Supergroup. We drive
out of the Gauteng, on a monotonous agricultural plateau covered by the Karoo Supergroup (350–200 Ma); coal is mined in this area
between Johannesburg and Witbank, and we
drive past some coal power plants. Near Witbank, the N12 merges with the N4 and after
150 km, we reach Middleburg toll plaza. We
drive another 100 km on the N4, before exiting
it at Machadodorp; we now are on the slightly
23
24
Day 1 : Saturday, 7 July
more rugged topography of the Transvaal Su- luzi, is Boesmanskop, made of 3.1 Ga syenites,
pergroup (2.4 – 2.1 Ga), which occasionally intrusive into the 3.45 Ga Stolzburg pluton.
crops out in small roadcuts.
60 km after Machadodorp, the R36 arrives
We leave Machadodorp on the R36; after ca. 20 at a T-junction; a low-lying ridge in front of
km, the road dips in the Komati valley through us is the Kees Zyn Doorns syenite, a 3.1 Ga
the Skurweberg pass. Some 15–20 km after the syenitic body related to the Boesmanskop compass, we reach the Archaean basement, immedi- plex. Turn left to Badplaas, which we reach
ately North of the small Kalkkloof Greenstone after another 5 km; 1.5 km in Badplaas, turn
right into the Aventura complex.
remmant (and mine), to the right.
The high grounds on either side of the valley are
made of the Black Reef quartzite, at the base
of the Transval Supergroup; the hills in front
of us, in the background, correspond to Barberton Greenstone Belt. The plain between us
and the Belt is made of 3.2 Ga TTG gneisses
from the Nelshoogte pluton. Driving SouthEast along the R36, we progressively see on our
right (South) hills corresponding to the Heerenveen and (further to the South-East) Mpuluzi
3.1 Ga potassic batholiths. A prominent hill
pointing out of the plain, in front of the Mpu-
(ca. 315 km, 3 hours plus stops).
Accommodation at the Aventura is in 4-sleepers
“rondavels” within the complex. Laundry facilities, hot and cold pool are available in the
Aventura – do not miss the hot pools at the
“hydro” spa (43◦ , 36◦ and 27◦ ). ATMs and limited shopping facility (superette, pharmacy and
liquor shop) are present near the Aventura reception. Breakfast and diner will be served at
the restaurant in Badplaas hotel (outside of the
main complex, near the R36 — not the “Gangster family restaurant” near reception).
GPS coordinates for all stops are expressed in degrees, minutes and decimals of minutes (WGS84).
Day 2: Sunday, 8 July
The lower part of the BGB and granite–greenstone contacts
S26◦ 02.549’; E31◦ 00.229’
Stop 2.1
Komati river section, Songimvelo nature reserve
The Kromberg and Hooggenoeg formations of the Onverwacht group at the base of Barberton Greenstone Belt
Access : Turn left out of the Aventura, on the R38 towards Machadodorp. After 1.5 km, turn right (east)
onto the R541, towards Mbabane/Tjakastad. Follow the road to eLukwatini crossroad (commercial zone),
24 km from Badplaas. Continue on the same road for another 27 km. After the end of the tar road, turn left
into eKulindeni village, drive past the village and reach the entrance gate to Songimvelo Nature Reserve after
3 km. Drive 3 km into the reserve to Kromdraai Camp. Leave the vehicles, and walk down 800 m to the
bridge on komati River, and ca. 2000 m upstream on the right (South) bank to the beginning of the section
(total 60 km).
Aim: Examine the stratigraphy and composition of the Onverwacht group, with special
reference to the H6 dacitic/detrical unit and its meaning.
Context: In Songimvelo Nature reserve, the
komati river cuts across low grade rocks from
the Onverwacht group, on the Eastern flank of
the Onverwacht anticline. The section we will
study starts near the top of the Hoogenoeg formation, and extends to the base of the Mendon
formation. The dominant lithologies are mafic
to ultramafic lavas (basalts and komatiitic
basalts), mostly pillowed; with some intercalations of cherts and detrical sediments. Some of
the chert layers contain micro-structures, that
have been interpreted as evidence for early life
(cyanobacteria Walsh, 1992). Near the top of
the Hooggenoeg formation, the H6 unit is made
of detrical sediments (conglomerates and volcanoclastic sandstones), in part silicified, and
dacitic lavas or porphyries; the H6 unit formed
at 3.45 Ga, exactly the age of the Stolzburg and
Theespruit plutons
scribed here (numbers refer to localities on the
map figure 12, not all are described here):
Loc. 1 (S26◦ 01.382’; E30◦ 59.259’): A sequence of mostly pillowed basalts, from the upper Hooggenoeg formation. We are located just
below the well-dated H6 unit (3445 Ga, Armstrong et al., 1990; Kröner et al., 1991), in unit
H5v (figure 13). The basalt unit is capped by
a thin chert horizons; close to the chert unit,
the basalts become silicified, and eventually are
transected by massive chert veins (best exposures are on the Northern bank).
Loc. 2 (S26◦ 01.454’; E30◦ 59.288’): Sandstones and conglomerates from the H6 unit.
The 3445 H6 unit is made of a ca. 170 m thick,
upward-fining sequence starting with conglomerates, and overlain by sandstones, turbidites
and eventually silicified shales (cherts). The
Site description: The map, log and descrip- clasts in the conglomerates are mostly (silicitions are from Hofmann et al. (2004); some fied) dacites, with minor sandstones, cherts and
localities of particular interest are briefly de- feldspar porphyries.
25
26
Day 2 : Sunday, 8 July
Figure 12: Geological map of the komati valley in Songimvelo Nature Reserve (Hofmann et al., 2004)
Loc. 4 and 5(S26◦ 01.711’; E30◦ 59.419’): Two
chert units (K1c1 and K1c2; the GPS coordinates probably correspond to K1c2) are intercalated in basalts and silicified lapilistone.
The K1c1 unit is correlated with the Buck Reef
Chert (Lowe and Byerly, 1999). Microfossils
and microbial mats are reported from the K1c2
unit at this locality (Walsh, 1992).
Loc. 6 (S26◦ 01.735’; E30◦ 59.444’): Wellexposed pavement of pillow basalts, showing
convex-upwards tops. We are ca. 500 m above
the bottom of the Kromberg formation.
Loc. 9 (100 m upstream from the weir): Ultramafic lapilistones are described by Hofmann
et al. (2004).
Between the weir and loc. 10: Fuschsitic,
silicified, schistose, strongly altered ultramafic
rocks occur as large blocs either side of the
road. They are made of lenticular, wavy bands
of fuschitic volcanic rocks and associated chert
in a brown-weathering carbonate matrix.
Loc. 10 and 11: The contact between the
Kromberg and Mendon formations occurs near
the footbridge. Two successive chert bands correspond to the “footbridge chert”, unit K3c;
they are exposed NW of the bridge. Zircons
from a lava layer in the chert were dated at
3334 ± 3 Ma (Byerly et al., 1996). Massive
komatiitic basalts (with intercalated, disrupted
layers of cherts) represent the base of the Mendon formation and are exposed immediately
downstream of the bridge.
Figure 9: Stratigraphic log of the Komati
River section (modified after Viljoen and
Viljoen, 1969c; Lowe and Byerly, 2003).
Locality 1. A sequence of pillow basalt
and minor massive basalt, a few hundred
metres thick, forms the uppermost part of
the Hooggenoeg Formation. The basalt
sequence is capped by a thin chert
horizon (H5c of Lowe and Byerly, 1999).
Pillow basalt containing abundant ocelli
and, commencing from c. 50 m below the
chert bed, becomes silicified upsection.
Silicification is generally associated with
a colour change from greenish grey to
light grey. This silicification is associated
with the replacement of igneous minerals
by quartz, carbonate and sericite, and an
increase in SiO2 and K2O (Viljoen and
Viljoen, 1969c; Byerly and Lowe, 1991).
Silicified basalt is transected by massive
black chert veins in the uppermost few
metres (Fig. 10a). Chert-veined basalt is
capped along a sharp contact by a c. 1 m
thick horizon (H5c) of massive to thinly
laminated black chert that is, in turn,
overlain by laminated grey chert (Fig.
10b). Grey chert contains normally
graded laminae with accretionary lapilli.
Microfossils in black chert have been
reported from this horizon (Walsh and
Lowe, 1985; Walsh, 1992). The chert
horizon and the underlying chert dykes
are best exposed on the northern river
bank.
Locality
The in
chert
is overlain
a sharp and planar contact by a c. 170 m thick,
Figure
13:2. Log
thehorizon
Onvewacht
group along
(Hooggeupward-fining
unit,
noeg,
Kromberg sedimentary
and Mendon sequence.
formations),This
along
the termed member H6 of the Hooggenoeg
Formation,
hasinbeen
correlated
with dacitic
komati
section
Songimvelo
Nature
Reservevolcanic
(Hof- rocks of the west limb of the Onverwacht
Anticline
mann
et al.,(Lowe
2004)and Byerly, 1999). The sequence starts with massive, poorly sorted, cobble
to boulder conglomerate (Fig. 10c). The clasts consist predominantly of silicified dacitic
volcanic rocks with minor carbonated volcaniclastic sandstone, grey and black chert, and
feldspar-porphyry, in a coarse-grained, carbonated sandstone matrix. Dacite clasts have been
dated at 3445±3 Ma (Kröner et al., 1991). This age is identical to ages obtained from
intrusive/extrusive dacitic rocks of H6 (Kröner and Todt, 1988; Armstrong et al., 1990).
Massive conglomerate is overlain by very thick beds of normally graded and massive
conglomerate and very coarse-grained sandstone, followed by massive and parallel-laminated
sandstone with minor intercalations of pebble conglomerate. The upper part of the sequence
19
27
28
Stop 2.2
S26◦ 09.279’; E30◦ 57.165’
Contact of the Steynsdorp pluton
Tectonic contact between the 3.5 Ga Steynsdorp TTG dome and overlying supracrustal rocks
Access : Regain the vehicles at the main camp. Drive out of Songimvelo reserve, back to the main (dirt)
road out of eKulindeni, and turn right (South). Keep on the same road until you reach the village of Vlakplaats
(9.5 km from eKulindeni, 6 km after the intersection with Badplaas road); turn right into the village, and
follow dirt tracks for about 4 km, into the valley. After the last ford, walk due South until you reach the
stream, and walk upstream for about 1 km, examining the outcrops in the river (total 20 km).
Aim: Introduce the oldest rocks exposed in and around the Barberton greenstone belt;
illustrate contact relationships between TTG’s and supracrustals along the southern margin
of the greenstone belt.
Figure 14: Geological map of the Steynsdorp pluton and surroundings (Lana and Kisters, in prep.), showing
lithologies and bedding or foliation trends. Image in the background is an orthorectified gray-scale aerial
photograph, provided by the Counsil of the Geosciences of South Africa. Right: lower hemisphere equal
area projection of poles to bedding in the Komati formation (i), schistosity in the Theespruit formation (ii),
gneissosity in the Steynsdorp pluton (iii) and sheeted granites (iv).
Context: The rocks of the Steynsdorp area
in the southern parts of the Barberton greenstone belt are the oldest rocks of the granitoidgreenstone terrain. U-Pb zircon ages for the
Steynsdorp pluton indicate an emplacement of
the mainly trondhjemitic rocks at 3509 Ma
(Kröner et al., 1996). Felsic, predominantly
metavolcanic rocks of the structurally overlying Theespruit Formation have yielded ages of
3530–3540 Ma. The cores of some inherited
zircons in the “Vlaakplaats granodiorite”, near
Vlaakplaats village we just drove past, have
yielded ages of ca. 3700 Ma, to date the oldest ages reported from the Barberton terrain
(Kröner et al., 1996).
From S to N, gneisses of the Steynsdorp
pluton are overlain by supracrustals of the
Theespruit Fm that comprise amphibolites, ultramafic talc-carbonate schists, characteristic,
up to 500m thick units of felsic quartz-sericite
schists and minor cherts, which, in turn, are
overlain by low-grade basalts and pillow basalts
of the Komati Fm. The latter show low strain
intensities and well-preserved primary textures.
The high-lying ground south of the Steynsdorp
dome is made up of granites of the ca. 3.1 Ga
Mpuluzi batholith.
Cross-cutting, intrusive contact relationships between the Steynsdorp dome and the
supracrustals are still preserved.
For the
most part, however, this contact has been
highly tectonized, and the strongly foliated
and transposed supracrustals are parallel to
the granitoid-greenstone contact and the welldeveloped solid-state gneissosity within the
29
Steynsdorp pluton. All of these fabric elements
have been folded into a shallow- to moderate
NE plunging antiform, that is parallel to the
main structural grain of the belt. A unidirectional, moderate NE-plunging mineral stretching lineation is pervasively developed throughout the gneisses and rocks of the Theespruit
Formation.
Fabric intensities generally decrease very
rapidly away from the granitoid-greenstone
contact, although strain is heterogeneous.
Similarly, metamorphic grades generally decrease from the amphibolite-facies, close to
the granitoid-greenstone contacts, to lowergreenschist facies grades within ca. 1.5 km of
the contact. Significantly, greenschist facies domains can be shown to alternate with variably
retrogressed amphibolite-facies domains over a
distance of ca. 1 km around the granitoidgreenstone contact.
Site description: The series of outcrops we will
visit close to the granitoid-greenstone contact
are dominated by amphibolites, garnetiferous
amphibolites, and felsic schists and agglomerates. In detail, the rocks contain pro (?)- and
retrograde assemblages that record very different metamorphic conditions. The outcrops
are interpreted to represent part of the southern extensional detachment that separates the
high-grade southern TTG terrain from the
overlying low-grade greenstone belt. Kinematic indicators are rare, but invariably point
to greenstone-belt-down, Steynsdorp pluton-up
movement.
30
Figure 15: a) Simplified lithological and structural map of the core of the Steynsdorp anticline. b) Sketch
of a NE-SW cross-section showing S1 in the three main lithological zones (sheeted zone, TTG gneisses and
supracrustals) in the core of the Steynsdorp Anticline. Note that there is a slight variation in dip angle from
the supracrutals to the Sheeted zone and that the horizontal Mpuluzi sheet is structurally overlying the NEdipping lit-par-lit sheets. Insets show S-C fabric with a normal sense of shear (circle) and stretched pillows at
the contact between the Theespruit and Komati Formations (rectangle). (Lana and Kisters, in prep.)
31
S26◦ 05.871’; E30◦ 57.532’
Stop 2.3
Contact of the Dalmein pluton
Sharp, intrusive contact of the Dalmein pluton with the Onverwacht group (time permitting)
Access : Drive back to the main (gravel) road; turn left (North), past Steynsdorp village. At the intersection,
turn left towards Badplaas. After ca. 4 km, the road dips to cross a stream (Dalmeinspruit), corresponding
to the contact of the pluton (total 14 km).
Aim: Examine the ca. 3.21 Ga Dalmein pluton, its undeformed nature and relations with
amphibolites from the BGB.
Context: The Dalmein pluton is a 3215 Ma
(Kamo and Davis, 1994) potassic, post-tectonic
pluton. It sharply truncates the structures in
the Southern Barberton Belt.
made of a much coarser variety, with 2–5 cm
K-feldspar phenocrysts and abundat, 10–80
cm MME. On the right bank of the stream
(East), the hill is made of amphibolites from
the Hooggenoeg formation; the actual contact
Site description: Pavements of the Dalmein
is obscured, but the undeformed character of
granite occupy the stream bed. It’s a slightly
the pluton a few meters from the contact (in
prophyritic granite, with microgranular mafic
the river bed) is striking.
enclaves (MME). The bulk of the pluton is
S26◦ 02.218’; E30◦ 48.115’
Stop 2.4
Deformed intrusive breccia of the Theespruit pluton
Intrusive contact relationships between the ca. 3.45 Ga Theespruit Pluton and the base of the
Barberton greenstone belt
Access : Drive West from the last outcrop for 18 km into eLukwatini. Turn right (N) into a dirt road ca.
100m before you cross the bridge across a small tributary river of the Theespruit River; continue with the
track for ca. 300m, staying close to the small creek. The outcrops are a number of large pavements in the
river showing intrusive breccias of amphibolites in trondhjemite (total 19 km).
Aim: Illustrate the magmatic assembly and later metamorphic overprint of the southern
high-grade terrane.
Context: The ca. 3.45 Ga Theespruit pluton is intrusive into amphibolite-facies rocks of
the lower Onverwacht group (Theespruit and
Sandspruit formations) of the N-S trending, socalled Tjakastad schist belt. Intrusive breccias point to the magmatic assembly of this
terrane at ca. 3.45 Ma. However, the age of
the amphibolite-facies metamorphism is only
ca. 3.23 Ga and related to the main collisional
event (D2 ). Garnetiferous amphibolites have
been used to constrain the metamorphic conditions of these amphibolites to 9 ± 1 kbar (van
Vuuren and Cloete, 1995); similar metamorphic
conditions of 7.5±1.0 kbar at T ca. 550-600 ◦ C
were obtained by Diener et al. (2005, 2006) for
metasediments to the immediate NW of these
outcrops.
Site description: These exposures illustrate
the intrusive contacts between the ca. 3.45
Ga trondhjemites of the Theespruit Pluton,
found to the East, and amphibolite-facies
supracrustals, here mainly amphibolites, of the
Theespruit and Sandspruit Formations. In plan
view, the intrusive breccias seem relatively undeformed, characterized by angular fragments
without any preferred orientation. In vertical
sections, however, fragments can be seen to be
strongly stretched. This constrictional strain
is typical for the 3.2 Ga strains related to the
oblique extrusion and exhumation of the highgrade rocks during the orogenic collapse of the
thickened belt.
32
Day 2: Sunday, 8 July
Bibliography
Armstrong, R., Compston, W., De Wit, M. J., and Williams, I. (1990). The stratigraphy of the
3.5-3.2 Ga Barberton greenstone belt revisited: a single zircon ion microprobe study. Earth
and Planetary Science Letters, 101:90–106.
Byerly, G. R., Kroner, A., Lowe, D. R., Todt, W., and Walsh, M. M. (1996). Prolonged magmatism and time constraints for sediment deposition in the early Archean Barberton greenstone
belt: Evidence from the Upper Onverwacht and Fig Tree groups. Precambrian Research,
78(1-3):125–138.
Diener, J., Stevens, G., and Kisters, A. F. (2006). High-pressure low-temperature metamorphism
in the southern Barberton granitoid greenstone terrain, South Africa: a record of overthickening and collapse of Mid-Archaean continental crust. In Benn, K., Mareschal, J.-C., and Condie,
K., editors, Archean Geodynamic Processes, volume 164 of monographs, pages 239–254. AGU.
Hofmann, A., Anhaeusser, C. R., Eriksson, K., and Dziggel, A. (2004). Excursion guide to
the geology of the Barberton greenstone belt. Technical report, Economic Geology Research
Institue, University of the Witswatersrand, Johannesburg.
Kröner, A., Byerly, G., and Lowe, D. (1991). Chronology of early Archaean granite-greenstone
evolution in the Barberton mountain land, South Africa, based on precise dating by single
zircon evaporation. Earth and Planetary Science Letters, 103:41–54.
Kröner, A., Hegner, E., Wendt, J., and Byerly, G. (1996). The oldest part of the Barberton
granitoid-greenstone terrain, South Africa: evidence for crust formation between 3.5 and 3.7
Ga. Precambrian Research, 78:105–124.
Lowe, D. and Byerly, G. (1999). Stratigraphy of the west-central part of the Barberton greenstone
belt, South Africa. Geological Society of America Special Paper, 329:1–36.
van Vuuren, C. and Cloete, M. (1995). A preliminary P-T-t path for the emplacement of the
Theespruit pluton, Barberton greenstone belt. In Centennial Geocongress, pages 315–318.
Geological society of South Africa.
Walsh, M. M. (1992). Microfossils and Possible Microfossils from the Early Archean Onverwacht
Group, Barberton Mountain Land, South-Africa. Precambrian Research, 54(2-4):271–293.
Day 3: Monday, 9 July
The high-grade Stolzburg terrane
The ca. 3.1 Ga Boesmanskop syenite
ca. S26◦ 00.333’; E30◦ 46.500’
Stop 3.1
Stolzburg pluton contact
Contact between the Tjakastad schist belt and the Stolzburg pluton
Access : Turn again left out of the Aventura and right on Lochiel road (R541). After 13 km, turn left
towards Tjakastad. The road travels across the Stolzburg pluton for some 9 km. The contact between the
Stolzburg pluton and the Tjakastad schist belt is marked by a dyke that forms a prominent approximately
N–S orientated ridge. Park just short of the dyke and walk through the field to the small koppie of Stolzburg
exposure to the south of the road. From these rocks walk over the dyke and down to the prominent pavements
below exposing the contact zone (total 21 km)
Aim: Demonstrate the complexities of the intrusion zones between the metamorphosed and
deformed amphibolite facies components of the lower stratigraphy of the greenstone belt and
the older TTG plutons.
Context: Where exposed, the contacts of the
older TTG plutons are complex as these rocks
have been metamorphosed and exhumed along
with the higher grade portions of the greenstone
sequence. The early interpretation of these
zones were that they formed by “dynamic” contact metamorphism as the plutons intruded as
diapirs. This was always somewhat at odds
with the evidence in other areas for shallow intrusion and the existence of intrusion breccias.
This has been reconciled by metamorphic and
geochronology studies that have shown that the
age of metamorphism is some 220 Ma younger
than the magmatic age for these rocks and that
metamorphism was syntectonic with the exhumation of the higher grade domains within
which these plutons occur. Thus, the contacts
between these plutons and the greenstone sequence record a complex history starting with
3.45 Ga intrusion and continuing through collision, burial, amphibolite facies metamorphism
and high temperature deformation associated
with their emplacement at higher crustal levels.
Site description: At this locality the Stolzburg
pluton presents itself as a relatively fine
grained, even textured trondhjemite. The contact trends north-south and is near vertical.
The foliation is parallel to the contact and
varies in intensity through the exposures. The
higher strain domains are relatively well defined
and increase in frequency approaching the contact. At the contact, the wall rocks are strongly
foliated felsic schists.
33
34
Day 3 : Monday, 9 July
Stop 3.2
S26◦ 00.174’; E30◦ 47.356’
Felsic agglomerates in the Tjakastad schist belt
Various components of the Theespruit formation, and metamosphism of the Tjakastad schist belt
Access : Regain Tjakastad road, continue East for about 1 km. The outcrops are accessed by turning off
the road to the right and driving some 200 m into the small valley between the two prominent hills. The
exposures are at the foot of the right hand side hill (total 2 km).
Aim: Illustrate (a) the volcanoclastic nature of parts of the Theespruit Formation, particularly the felsic-schist component, and (b) the occurrence of low-strain domains in the
otherwise highly strained sequence characterized by pervasive prolate strains.
Context: The Theespruit Formation is characterized by abundant felsic schist, besides
amphibolites and ultramafic serpentinites and
talc-carbonate schists. Commonly fine-grained
and finely-laminated, there are numerous localities where the felsic schists are developed as
agglomerates, containing fragments of up to 20
cm in diameter, underlining the originally pyroclastic and/or volcanoclastic nature of the felsic
schists. Most agglomerates are monomict.
Site description: This outcrop presents one
such agglomerate at relatively low strain intensity. Several layers of deformed felsic agglomorate are exposed at the base of a prominent
hill, consisting mostly of serpentinised dunite.
These rocks are deformed with clasts extended
to define prominent near vertical rods. How-
ever, this is a relatively low strain domain
within Theespruit formation, which has generally been deformed to the the point where
such primary features are almost impossible to
recognise. Original bedding is well-preserved
and fragments are only slightly elongated. Typically, pyroclastic fragments are rodded due to
the pervasive prolate strains that have affected
the Theespruit Formation. The agglomerates
and the fine-grained matrix consist primarily of
quartz, plagioclase, muscovite with occasional
chlorite. On the far side of the road, felsic
schists can be viewed that carry more typical
fabrics. In these rocks, the volcaniclastic origin
is hard to determine and the mineralogy consists of quartz + muscovite ± chloritoid (posttectonic).
J.F.A. Diener et al. / Precambrian Research 143 (2005) 87–112
35
91
Fig. 2. Geological map of the N–S trending Tjakastad schist belt that is bordered by the Stolzburg and Theespruit plutons. The position of sample
localities is also shown.
Figure 16: Geological map of the Tjakastad schist belt and its contacts (Diener et al., 2005).
mylonites are characterized by strongly prolate (L S)
fabrics and L1 is the strongest fabric element developed in these rocks (Fig. 3b). Clasts and mineral aggregates have axial ratios of 20–100+:1–3:1, indicating that
fabric development occurred in a highly constrictional
strain regime. In the northern part of the study area,
the strongly prolate amphibolite-facies mylonites grade
into, and are overprinted by, greenschist-facies mylonites
away from the plutons and towards the central parts of
the BGB. Strain markers in these lower grade mylonites
36
Stop 3.3
S25◦ 58.378’; E30◦ 50.208’
Komatiite type locality
Komatiites lava flow near the type locality
Access : From the previous stop, drive back to main road, turn right (east) and drive into Tjakastad.
Continue with this road/dirt track and cross the bridge over the Komati River (5 km). Turn right (east) after
ca. 100 m after the bridge and onto dirt-track and continue for ca. 1.4 km. Stop alongside the road, and
walk down to a small river to the right (east) of the road for ca. 100m. Komatiite outcrops are in the river
valley. Please do not use your hammers on these outcrops; there will be an opportunity to sample the
komatiites, spinifex textures etc. just after these exposures (total 6.5 km).
Aim: Observe (a) typical komatiites (b) the low metamorphic grade of the rocks of this
portion of the greenstone belt.
Context: The Komati formation, of the Onverwacht group, is characterized by the presence of numerous komatiites flows. Komatiites
were defined by Viljoen and Viljoen (1969), in a
stream a few kilometers from here; the original
outcrop is overgrown and difficult to access.
These outcrops are located to the north of the
Komati Fault close to the base of the Onverwacht Group in rocks of the Komati Formation. The rocks have undergone low-grade,
greenschist-facies metamorphism (T: 350 ±
50◦ C; P: 2.6 ± 0.6 kbar Cloete, 1999). The
low metamorphic grades are well demonstrated
by the metamorphosed inter-pillow fill material
Stop 3.4
exposed with the pillow lavas, which consists of
chlorite and epidote.
Site description: At this locality komatiites
can be viewed illustrating well-layered, lowstrain and low-grade metamorphic komatiite
flows with olivine cumulate zones, spinifex textures and breccia flow tops. The stratigraphic
high is to the North (upstream). Some 50 m
upstream form this locality very well developed pillow lavas are exposed. Another 4 km
along the same road takes you to a small road
side exposure of rubble and loose blocks of nice
spinifex-textured komatiite.
S25◦ 59.844’; E30◦ 39.888’
The ca. 3.2 Ga Inyoni shear Zone (ISZ)
Contact zone between the ca. 3.2 Ga Badplaas terrane and the ca. 3.45 Ga Stolzburg terrane
Access : Return to Tjakastad and retrace the road towards Badplaas. Turn right onto the R541; after 1.5
km, turn left (South) on a dirt track opposite a sign “Inyoni”. Drive 1.2 km to the South and stop opposite
a small gate in the fence; cross the fence and walk West ca. 100 m (total 20 km).
Aim: Examine the rocks, structure and metamorphism in an Archaean terrane boundary.
Context: The Inyoni shear zone is a ca. 2
km wide arcuate gneiss belt, that separates the
3.45 Ga Stolzburg pluton, to the East, from the
complex ca. 3.2 Ga Badplaas gneiss terrane, to
the West. The Northern extent of the Inyoni
sher zone can be traced into the Stolzburg arm
of the Barberton Greenstone Belt; its southwestern extremity is deflected by, and eventually merges with, the ca. 3.1 Ga Welverdiend
Shear Zone, a synmagmatic shear zone control-
ling the emplacement of the Mpuluzi batholith
(see day 5).
The gneisses of the ISZ display multiple intrusive relationships among a variety of grey,
trondhjemitic banded gneisses, ranging from
pre-, to syn-, to slightly post-tectonic. They
contain a variety of inclusions of all size (10
cm – 100 m) of supracrustal rocks (dominantly
amphibolites, but minor clastic sediments and
37
No
BIF are present), metamorphosed up to upper strike, with a subvertical or steep dip to the
amphibolite facies.
East. A weak, vertical stretching lineation,
parallel to the axes of open folds affecting
the main foliation, is sometimes observed, but
Coarse-grained, leucocratic
shear sense indicators are rare and conflictsyntectonic intrusion
ing. The lineation has the same orientation
Supracrustal package, dominated
as the dominant (solid-state) stretching linby amphibolites; with minor B.I.F.,
metasediments and ultramafics.
eation within the Stolzburg domain (stops 3.1
Supracrustal package, dominated
to 3.3), defining a coaxial vertical stretching
by metapelites; with minor
amphibolites and ultramafics.
fabric interpreted as an exhumation-related deformation. Rare evidence for older deformation
is preserved in supracrustal inclusions, in the
form of isoclinal folds with a shallow axis and
(rare) sub-horizontal lineations.
f.
ou
o
tcr
rg p
luto
n
No outc
rop
p
No
ou
tcr
op
Sto
l zb u
Badplaas
Gneisses
No
ou t
cro
p
Lines
F. axis
Lineation
Poles to planes
F. plane
Dextral
Sinistral
Foliation
Figure 17: Geological map of the ISZ. Gneisses are
left white; the thin lines denote foliation trends. The
star correspond to the visited localities. The two
stereograms below show the lineations and fold axis
(left) and the poles to foliations, shear zones and fold
planes (right); the star and the thick line correspond
to the pole of the best plane.
A pervasive, solid-state to migmatitic foliation is observed in both the gneisses and the
supracrustals. It shows a consistent northerly
Moving across the Inyoni Shear Zone into the
Badplaas domain, the vertical lineation disappears; the dominant structure becomes a vertical foliation, with crenulation folds and conjugate shear zones corresponding to East–West
coaxial flattening. Over a distance of 2–3 km,
the shallow southeasterly dipping (ca. 135/30◦ )
gneisses of the Badplaas domain are steepened
into the subvertical gneissosity of the ISZ. The
western part of the ISZ also becomes progressively swamped by syn- to post-tectonic intrusions of various plutonic phases that can
be traced in continuity into the Badplaas terrane.
The supracrustals from the ISZ are mostly
hornblende–epidote–plagioclase amphibolites
with little potential for metamorphic studies.
Occasionally however, some useful assemblages
are exposed, including clastic metasediments
(diopside/hornblende + andesine + quartz ±
garnet), meta-BIF (quartz + ferrosilite + magnetite + grunerite) and garnet-bearing amphibolites (hornblende + epidote + plagioclase + garnet ± clinopyroxene ± quartz).
The different lithologies , using a variety of
geothermometers and geobarometers, as well
as thermocalc estimates, give peak conditions between 650◦ C and 800◦ C for pressures
ranging from 8 to 11 kbar (Dziggel et al., 2002;
Moyen et al., 2006), corresponding to apparent
geotherms of less than 25 ◦ C.km−1 . Leucosomes in some amhibolites (associated with
either garnet or amphibole) point to melting
reactions such as hornblende + plagioclase +
quartz + H2 O → (tonalitic) melt + garnet
(Moyen and Stevens, 2006) or hornblende +
38
plagioclase+quartz+H2 O → (tonalitic) melt +
amphibole2 (at high water activities: see Gardien et al., 2000); both reactions occuring in
the same temperature range. The metamorphic minerals define the main fabric, and the
melt patches are syntectonic, showing that this
metamorphism corresponds to the main deformation phase.
Elements of a prograde metamorphic history
are recorded in the core of zoned garnets, or
in garnet growth sites. There, coexisting albitic plagioclase and calcic garnet point to relatively high pressure, low temperature conditions; indeed, thermocalc estimates indicate P–T conditions corresponding to a 15–
18◦ C.km−1 . apparent geotherm —absolute values for P and T are less precise, but are somewhere between 8 and 15 kbar, 550 and 700
◦
C.
The timing of deformation (and the peak metamorphism) in the Inyoni Shear Zone is constrained by U-Pb zircon age of 3229 ± 5 Ma
obtained on a late-tectonic trondhejimitc dyke
(Dziggel et al., 2002, 2005), consistent with a
metamorphic sphene age of 3229±9 Ma in clastic metasediments from the shear zone.
partially molten amphibolite. Here, the leucosomes are amphibole-rich, and form a mush of
a very leucocratic melt associated with large,
euhedral amphiboles, probably peritectic products of the melting reaction.
3.4.d (S25◦ 59.750’; E30◦ 39.699’): Near the
slope break South-West of the previous locality, large, rounded pavements are made of the
Inyoni gneisses. Several phases are observed,
including an older, relatively coarse-grained variety (possible equivalent of the Stolzburg pluton?), and finer-grained, syn-tectonic phases
similar to the dykes dated 3229±5 Ma (Dziggel
et al., 2005). Rare, dilacerated elements of
amphibolites are also observed (they are more
spectacular at the next locality), as well as a
large cutting dyke of white granite, possibly related to the ca. 3.1 Ga magmatism. The vertical foliation, refolded by vertical axis, open
folds, is evident here.
3.4.e (S25◦ 59.896’; E30◦ 39.673’): Walk South
(left), more or less level, and cross the gully. It
is overlooked (on the left/South bank) by outcrops of the composite Inyoni Gneisses; here,
the gneisses are rich in small enclaves of amphibolite, occasionally garnet-bearing. Very similar amphibolites some 500 m further South
show nice textures of simultaneous growth of
albite and garnet (samples INY 131, INY 132),
that were the base of the metamorphic interpretation in Moyen et al. (2006). The vertical-axis
folds are also evident in this locality.
Site description: Outcrop in this part of the
Inyoni Shear zone is scattered, but quite abundant; we’ll walk in the shear zone, and to different localities. The track we’ve been using
runs along the Eastern margin of the ISZ, on
poorly exposed, high strain gneisses from the
3.4.f (S25◦ 59.987’; E30◦ 39.698’): Walk back
Stolzburg pluton.
up (East) towards the dirt track and the cars.
◦
◦
3.4.a (S25 59.834’; E30 39.813’): Walk to an Underway just above the slope break, it is posisolated tree at the head of a gully, some 200m sible to observe many small outcrops of various
of the gate. Pavements under the tree show metamorphic rocks, including a most unspecthe dominant, epidote-bearing amphibolite lit- tacular, meter-sized boudin of garnet-bearing
par-lit injected by granitic dykes probably cor- amphibolites that probably occurs in the hinge
zone of a refolded isoclinal fold. This is the
responding to the end of the exhumation.
locality were we found the first evidence for
3.4.b (S25◦ 59.803’; E30◦ 39.799’): Walk to high pressure (> 12 kbar) metamorphism, in
your right (North); within a few tens of me- the core of garnets (sample INY 121).
ters, you’ll be walking among scattered fragments of metamorphic rocks, including BIFs,
ultramafic lithologies and metasediments, corresponding to the rocks studied by Dziggel et al.
(2002).
3.4.c (S25◦ 59.712’; E30◦ 39.772’): In and
around a group of trees are scattered blocks of
39
a
5
0
0
m
b
1
0
0
Figure 19: Summary of P–T estimates obtained from
the Inyoni Shear Zone. Hatched fields: metasediments (classical thermobarmetry Dziggel et al., 2002);
grey ellipses: amphibolites (thermocalc; Moyen
et al., 2006)
m
c
1
0
0
m
Figure 18: Metamorphic textures in Inyoni Shear
Zone amphibolites. A: Growth of euhedral garnet in
an albitib moat. B: Amphibole–quartz symplectites
surrounding garnet; symplectites formed during garnet breakdown. C: Relict of a relatively sodic amphibole in a garnet growth site. A and C are from samples
INY131 and 132, 500m South of locality 3.4.e; B is
from sample INY25, near site 3.4.c
40
S26◦ 01.973’; E30◦ 39.725’
Stop 3.5
Western slopes of Boesmanskop
Syenites and Syenogranites of the 3.1 Ga Boesmanskop pluton
Access : From the previous outcrop, continue with track to the S and aim for the western slopes of the
topographically prominent Boesmanskop Pluton; there is a network of tracks that will get you there (total 4
km).
Aim: Illustrate the syenites and syenogranites of the Boesmanskop pluton, part of the
late-stage, ca. 3.1Ga GMS suite in the Barberton terrain.
Context:
The Boesmanskop syenogranite complex, in
short referred to as the “Boesmanskop syenite” (Anhaeusser et al., 1983), forms part of
the 3.1 Ga GMS suite of potassic plutonic
rocks. The main outcrops of the Boesmanskop syenite underlie two very prominent hills
to the immediate north of the escarpment of
the Mpuluzi batholith. The syenogranites and
syenites of the Boesmanskop are typically reddish to pinkish in colour. They contain Kfeldspar, hornblende, and biotite ± plagioclase ± sphene ± quartz. The syenogranites and syenites typically display cumulatelike textures consisting of mm-sized, euhedral
K-feldspar and interstitial hornblende and/or
biotite. Recent mapping has shown that the
syenogranites have a far wider distribution than
previously thought. Gneissose varieties of the
syenogranites are exposed along the northwestern and western escarpment, where rocks of
the Boesmanskop syenite are caught up in the
synmagmatic Welverdiend Shear Zone (Westraat et al., 2004). The Boesmanskop pluton
is the largest of the syenitic to syenogranitic
intrusions. U-Pb zircon ages point to an
emplacement age of 3107+4
−2 Ma, within er-
ror of the emplacement ages of the granitic
batholiths of the adjacent Mpuluzi batholith or
the very large Nelspruit batholith in the north.
Compositionally and texturally very similar
rocks occur as the smaller, dyke-like Kees Zyn
Doorns syenite near Badplaas, along the western margin of the Mpuluzi batholith (stop 5.5)
and in the southern parts of the Heerenveen
batholith (stop 5.2), also emphasizing the genetic relationship between the distinct syenites/syenogranites and the large and composite
granitic batholiths of the GMS suite.
Site description: In these outcrops, a very
coarse, porphyritic textural variety of the Boesmanskop pluton can be seen that is confined
to the western flank of the pluton.
The
rock consists of K-feldspar (microcline, orthoclase and perthite) and plagioclase (oligoclasealbite). The main ferromagnesian mineral is
a blue green hornblende with, in places, cores
of green augite. Biotite is also present, quartz
is not common, but does occur, and euhedral sphene and apatite are very prominent in
places, whereas zircon and magnetite are accessories. Chlorite, epidote and sericite are secondary minerals.
41
S26◦ 03.442’; E30◦ 39.312’
Stop 3.6
Basal (?) contact of Boesmanskop pluton
Contact relationships between rocks of the 3.1 Ga Boesmanskop syenite and older TTG basement
gneisses and amphibolites
Access : From the previous outcrop, continue to the south for approximately 2 km, heading for a bridge
across the Theespruit River. Park the car before the bridge, cross the river and turn to your left immediately
after the gate, towards the main body of the Boesmanskop, staying close to the river. After ca. 250 and ca.
400 m, there are two large pavements in the river, exposing mainly older TTG’s and amphibolites. These
rocks are intruded by subhorizontal sheets and cross-cutting dykes of the Boesmanskop syenite (southern
pavements) and subhorizontal sheets of a greyish subvolcanic phase (northern platform) (total 3 km).
Aim: Illustrate the subhorizontal, sheet-like nature of many of the 3.1 Ga potassic intrusives.
Context: The Boesmanskop syenite is intrusive into the “basement gneisses”, mostly of the
Stolzburg pluton. Here, the basement is made
of banded gneisses belonging to the ca. 3.1 Ga
Welverdiend Shear Zone.
Site description: The Boesmanskop syenites and syenogranites are well-exposed in
pavements along the Theespruit River. The
syenogranites sharply truncate the underlying, steeply-dipping banded TTG gneisses and
amphibolites along a subhorizontal, slightly
undulating contact. The syenogranites appear macroscopically undeformed, displaying
the cumulate-like textures typical of the main
body of the Boesmanskop syenite to the immediate north. Small, meter-scale, often pegmatitic offshoots from the subhorizontal sheets
penetrate into the underlying amphibolites and
TTG gneisses, where they appear locally folded
with the steep foliation. This suggest ongoing deformation along the basement TTG’s
and amphibolites. A possible feeder dyke to
the subhorizontal syenogranite sheets is exposed along the eastern banks of the Theespruit
River. This dyke is also deformed, but connects
into undeformed syenogranites.
S26◦ 03.410’; E30◦ 39.421’
Stop 3.7
Hypovolcanic facies of the Boesmanskop syenite
Access : Continue downstream (towards the NE) from the previous stop on the eastern banks of the
Theespruit River for ca. 100 -150 m to a large pavement in the Theespruit River with a small waterfall.
Aim: Occurrence of a sheet-like subvolcanic phase related to the Boesmanskop syenite,
indicating a near surface position of the granite–gneiss terrain at ca. 3.1 Ga.
Context: Relatively rare subvolcanic phases
associated with the Boesmanskop syenite are
present. Their presence points to the near surface magmatic activity related to the 3.1 Ga
GMS suite in the region and may indicate that
the exhumation and possibly peneplain formation of the granite-gneiss terrain might have
been completed already at 3.1 Ga, 500 Ma before the deposition of the Black Reef Quartzite
at the base of the Transvaal Supergroup.
Site description: This outcrop was discovered
during the regional mapping of the southern
granite gneiss terrain in 2002–2003. A subhorizontal, very fine-grained, greyish sheet-like
phase sharply truncates TTG gneisses and amphibolites. The fine-grained intrusive phase
consists of plagioclase, K-feldspar, hornblende
and minor quartz. It contains 5 to 15 mmsized glomerophyric monomineralic aggregates
of plagioclase and hornblende. U-Pb zircon
42
Day 3: Monday, 9 July
ages indicate an age of 3096 ± 4 Ma for this
phase. This is slightly younger than the 3107+4
−2
Ma age for the bulk of Boesmanskop syenite to
the immediate north (Kamo and Davis, 1994)
and the 3113 to 3105 Ma Mpuluzi batholith
to the immediate southeast (Kamo and Davis,
1994; Westraat et al., 2004). The sheet intrudes along a slightly undulating, subhorizon-
tal surface and sharply truncates the underlying, steeply inclined TTG gneisses and amphibolites. There are several occurrences of dykes
that underlie the sheets. The dykes commonly
contain a strong solid-state fabric (both foliation and lineation), whereas the sheets appear
undeformed.
Bibliography
Anhaeusser, C. R., Robb, L. J., and Barton, J.M., J. (1983). Mineralogy, petrology and origin f
the Boesmanskop syeno-granite complex, Barberton mountain land, South Africa. Geological
Cloete, M. (1999). Aspects of volcanism and metamorphism of the Onverwacht group lavas in
the southwestern portion of the Barberton greenstone belt, volume 84 of Memoirs. Geological
Survey of South Africa.
Dziggel, A., Armstrong, R., Stevens, G., and Nasdala, L. (2005). Growth of zircon and titanite
during metamorphism in the granitoid-gneiss terrain south of the Barberton greenstone belt,
South Africa. Mineralogical Magazine, 69:1021–1038.
Dziggel, A., Stevens, G., Poujol, M., Anhaeusser, C., and Armstrong, R. (2002). Metamorphism
of the granite-greenstone terrane South of the Barberton greenstone belt, South Africa: an
insight into the tectono-thermal evolution of the ’lower’ portions of the Onverwacht group.
Gardien, V., Thompson, A., and Ulmer, P. (2000). Melting of Biotite+Plagioclase+Quartz
gneisses: the role of H2O in the stability of amphibole. Journal of Petrology, 41:651–666.
Moyen, J.-F. and Stevens, G. (2006). Experimental constraints on TTG petrogenesis: implications for Archean geodynamics. In Benn, K., Mareschal, J.-C., and Condie, K., editors,
Archean geodynamics and environments, volume 164 of monographs, pages 149–178. AGU.
Moyen, J.-F., Stevens, G., and Kisters, A. F. (2006). Record of mid-Archaean subduction from
metamorphism in the Barberton terrain, South Africa. Nature, 443:559–562.
Viljoen, M. and Viljoen, R. (1969). The geology and geochemistry of the lower ultramafic unit of
the Onverwacht group and a proposed new class of igneous rocks. Geological Society of South
Africa Special Publication, 2:55–86.
16.
Day 4: Tuesday, 10 July
The ca. 3.2 Ga plutons North-West of BGB
ca. S26◦ 04.750’; E30◦ 22.067’
Stop 4.1
Rooihoogte pluton
Polyphased TTG gneisses of the Badplaas/Rooihogte domain
Access : From Aventura (Badplaas), turn right and drive 23 km west along the R38 towards Carolina. Take
the small gravel road to your left demarcating “Sappi: Rooihoogte Forestry” and enter the pine forest. Follow
the road for approximately 2.5 km down into the valley passed the old farm buildings (Buffelspruit farm) to
the Buffelspruit River. Turn right along the forestry track and follow the river for approximately 1 km. Large
pavement outcrops in the river to the left and the slopes in to the forests to the right should be visible (total
25 km).
Aim: Document the polyphased nature of rocks of the Badplaas terrane.
Context: The Badplaas terrane in the SW
of the Barberton granitoid greenstone terrain
is compositionally and structurally the most
heterogeneous TTG terrane surrounding the
greenstone belt. Its Southern and Western extents are overlain by younger sediments or obscured by the intrusion of younger, ca. 3.1
Ga granites. The eastern contact with the
Stolzburg terrane is tectonic, marked by the
Inyoni Shear Zone, whereas it shows intrusive
relationships with the Nelshoogte pluton in the
North.
of the Badplaas terrane are invariably gneissose and the folding and refolding of gneissosities and compositional layering testify to the
polyphase deformation history of the terrane,
which is not recorded in rocks of the adjacent
and older Stolzburg terrane, nor in the Kaap
Valley terrane to the immediate north.
To date, no geochronological data have been
available from this part of the TTG-gneiss terrain. Recent age determinations by Kisters et
al. (in prep) from five of the main phases of the
Badplaas terrain indicate a range of ages of between ca. 3290—3230 Ma. Importantly, these
ages indicate some 60 Ma of TTG plutonism
prior to the main collisional phase (D2 ) at ca.
3230 Ma.
For the most part, the Badplaas terrane
represents a regional scale intrusive breccia,
showing a multitude of intrusive trondhjemite
phases. Shallowly-dipping, banded and transposed trondhjemitic gneisses dominate the
southern parts of the terrane, east of the
younger (3.1Ga) Heerenveen batholith. The
western extent of the Badplaas terrane, formerly referred to as the Rooihoogte pluton
(Anhaeusser et al., 1983), shows at least three
distinct and regionally widespread gneissose
trondhjemite phases. Coarse-grained biotite
leuco-trondhjemites, previously referred to as
the Batavia phase, are developed along the
eastern boundary of the Badplaas terrane, bordering against the Inyoni shear zone. Rocks
Site description: Two samples were taken for
geochronological work from this outcrop of the
composite Rooihoogte Pluton. The light-grey,
strongly gneissose trondhjemite is a regionally
widespread phase. It is intruded and brecciated
by a dark-grey phase. U-Pb zircon ages from
the light grey phase indicate an emplacement of
3291+19
−13 Ma, whereas the younger phase yields
an age of 3258±13 Ma. This also constrains the
timing of the fabric-forming event in the main
trondhjemite phase to > ca. 3260 Ma.
43
44
Day 4 : Tuesday, 10 July
Stop 4.2
S25◦ 54.867’; E30◦ 39.083’
The Nelshoogte pluton
Tonalites and trondhjemites of the Nelshoogte pluton
Access : Drive back to Badplaas (25 km). Take the main road from Badplaas to Barberton (R 38) and
turn right (E) after ca. 10 km into a dirtroad; continue with this track for ca. 4.5 km which will bring you
to the Komati River (total 40 km).
Aim: Illustrate the relation between syntectonic (D2 ) trondhjemites and late- to posttectonic tonalites of the Nelshoogte Pluton.
Figure 20: Geological map of the Nelshoogte pluton (Belcher et al., 2005). Five phases are mapped — Two
early trondhjemites, three late tonalites.
Context: The oval-shaped Nelshoogte pluton
has dimensions of approximately 17 by 15 km
forming a topographic depression, surrounded
by the elevated topography of the Barberton
Supergroup, mainly Onverwacht Group rocks,
in the north and east, and by the Transvaal Supergroup in the northwest and west. It shows
intrusive contacts with the Onverwacht Group
and is unconformably overlain by the Transvaal
Supergroup. Various ages have been obtained
for the pluton suggesting a protracted emplacement history of between ca. 3236 and ca. 3224
Ma.
with a variably developed gneissosity and, volumetrically subordinate, weakly to non-foliated
tonalite. Intrusive relationships between the
different phases are rare, but where exposed
indicate the tonalitic phases are younger than
the trondhjemitic phases. The trondhjemites
are made up of quartz, plagioclase, biotite, and
accessory muscovite, sphene, zircon, and apatite with secondary sericite, epidote and calcite, mainly related to the saussuritization of
plagioclase. Tonalites are commonly greyish,
medium- to coarse-grained comprising plagioclase, quartz and hornblende, occasional biotite
and microcline, accessory zircon, apatite and
The Nelshoogte Pluton consists of relatively hosphene.
mogeneous leuco-trondhjemite to trondhjemite
45
Figure 21: Strain pattern and intensity in the Nelshoogte pluton (Belcher et al., 2005). Darker shade
correspond to higher strain intensity.
A pervasive solid-state foliation is developed
across the pluton within the trondhjemitic
phases being defined by disc-like quartz aggregates and quartz ribbons, the grain-shape
preferred orientation of recrystallized quartzfeldspar aggregates and the preferred orientation of biotite and/or hornblende. An associated stretching lineation defined by muscovite,
biotite and/or hornblende plunges at moderate angles to the southeast. The gneissosity
has a consistent NW-SE trend over much of
the pluton, but rotates anticlockwise towards
the margin of the pluton and the contacts with
the surrounding greenstones, so the gneissosity
is largely parallel to the pluton-wallrock contact. The marginal phases of the pluton are
strongly gneissose and show, in places, protomylonitic textures. The tonalitic phases crosscut the gneissosity and are weakly foliated to
undeformed.
Site description: This outcrop shows different
trondhjemitic phases that are, on a meter-scale,
intruded by coarse-grained tonalitic phases
along the banks of the komati River.
46
Stop 4.3
S25◦ 51.200’; E30◦ 45.477’
Nelshoogte pass
Intrusive breccias of the Nelshoogte pluton in Onverwacht Group amphibolites (time permitting)
Access : Drive back to the tar road (R38); turn right and drive for ca. 15 km. Stop in some high road cuts
near the top of the pass (total 20 km). Be extremely careful when examining the outcrop, located in a
blind corner.
Aim: Document the brittle intrusive contact of the Nelshoogte pluton in the surrounding
greenstones.
Context: The Nelshoogte pluton shows generally lit-par-lit intrusion relations with the surrounding greenstone lithologies. However, very
brittle, intrusive breccias are also recorded.
Stop 4.4
Site description: The contact appears here as
a magmatic breccia, with 10–100 cm angular
clasts of amphibolite in a matrix of fine grained
tonalite/trondhjemite from the Nelshoogte pluton.
S25◦ 45.823’; E30◦ 49.426’
Northern side of Nelshoogte pass
Viewpoint on the Kaap Valley
Access : Drive on the R38 for another 15 km; the roads crosses a forested plateau and dips down into the
Kaap Valley. Shortly after a sharp right bend, stop on little parkings on the left hand side of the road, facing
high cliffs (road cuts) on the right hand side (total 15 km).
Aim: Observe the typical hornblende-tonalite from the Kaap Valley pluton, and discuss
the unique nature of this pluton in the BGGT.
Context: The Kaap Valley Pluton is the largest
of the TTG plutons, and the only large tonalitic
pluton in the BGGT (the other large plutons
are trondhjemitic)(Robb et al., 1986; Faure
and Harris, 1991). U–Pb zircon ages indicate
an age of ca. 3227 ± 2 Ma for its emplacement (Kamo and Davis, 1994), pointing to its
syn- to late D2 emplacement. Smaller tonalite
bodies with similar emplacement ages are also
present in the south of the granitoid-greenstone
terrain, north of the Schapenburg Schist Belt
(Stevens et al., 2002; Anhaeusser, 1983) and in
the Nelshoogte Pluton to the immediate south
(stop 4.2).
The Kaap Valley pluton is made of a very
homogeneous, medium-grained amphiboletonalite (to leuco-diorite). Microgranular mafic
enclaves are common, and the rock is often weathered and (secondary) chlorite- and
epidote-rich.
Site description: The topographic low in front
of us is occupied by the generally strongly
weathered Kaap Valley pluton. The flat lying
summits on the left and front are the Transvaal
Supergroup cover; the Barberton greenstone
belt is on the far right. The outcrops on the
other side of the road allow to examine the
tonalite.
47
S25◦ 46.150’; E31◦ 03.700’
Stop 4.5
Border of the Kaap Valley pluton above Barberton town
Contact between the Kaap Valley Pluton and the BGB
Access : Follow the R38 down into the Kaap Valley for 22 km; leave Nelspruit road on your left, continue
for another 7 km to a large 4-way stop, (the Barberton Mediclinic is located on the corner). Continue straight
on the R40 towards Josefsdal; the road takes you into the greenstone belt for ca. 2.5 km. After a sharp
serpentine, the contact between the Kaap Valley Pluton and the greenstone belt is exposed on a long, gently
climbing straight. A parking lot is situated some 500m further up the road on the left-hand side (total 31
km).
Aim: Observe (a) the sheared nature of the contact of the Kaap Valley Pluton; (b) the low
metamorphic grande of the enveloping greenstones.
Context: Like the other ca. 3.2 Ga plutons,
the Kaap Valley tonalite is intrusive into the
greenstone belt. In this case however, the contact is sheared, consitent with syn-exhumation
emplacement.
greenstone belt. Notably, the high-grade metamorphic basal greenstone sequence is not developed here, although amphibolite-facies rocks
are developed further to the north and south
of this contact. The contact is steep to overturned, dipping to the East. The tonalites conSite description: In these outcrops, rocks of the
tain a variably developed solid-state gneissosity
Kaap Valley Pluton are intrusive into low-grade
parallel to the contact.
metamorphic supracrustals of the Barberton
S25◦ 47.517’; E31◦ 05.049’
Stop 4.6
Moodies conglomerates along the R40
Conglomerates and sandstones of the Moodies group
Access : Drive another 6 kilometers along the R40, until you reach a 90◦ turn to the left. Stop immediately
before and examine outcrops on the right-hand side of the road (total 6 km).
Aim: Observe the detrical sedimentation of the Moodies group, and its implications for
orgenic evolution in the BGGT; discuss the origin and meaning of granitic clasts in the
Moodies conglomerate.
Context: The ca. 3.21 Ga Moodies group is
the upper-most sedimentary unit in the Barberton Belt. It is made of sandstones and conglomerates, occuring in discontinuous, probably fault-bounded basins (Heubeck and Lowe,
1994b,a) during the D2b exhumation. The
Moodies group effectively represent molassic
deposits following the 3.2 Ga orogeny. The
clasts are made of all the lithologies occuring in and around the belt (cherts, metamafic
lithologies, granitoids. . . ), but recent sampling
in outcrops on the Northern part of the belt
revealed a population of granitic clasts corresponding to rocks of age (Kröner and Compston, 1988) and nature unknown in outcrop-
ing rocks. Some of the oldest clasts (> 3500
Ma) appear to be potassic granites (Kröner and
Compston, 1988), suggesting an already stabilized crust at this age.
Site description: The outcrop is made of alternating sandstones and clay-rich levels. Each
sandstone-clay pair is well sorted, with a conglomeratic base fining upwards to a clay-rich
top. A similar feature is observed at the
outcrop scale, with the conglomerates becoming progressively coarser to the right (NW, or
stratigraphic down) of the outcrop. Crossbeddings and polygonal dessication cracks are
obvious in the outcrop.
48
S25◦ 47.204’; E31◦ 04.996’
Stop 4.7
Panorama on the Fig Tree Valley
Structural panorama on the BGB and the Inyoka fault system
Access : Drive back down towards Barberton; stop in a hairpin about 1 km from the previous stop.
Aim: Document the deformation style within the BGB and discuss the nature and importance of the Inyoka Fault.
Context: All the lithologies of the BGB are
folded together, in a tight, isoclinal style. Internal unconformities are present, but not immediately obvious; the Moodies group is clearly
discordant on the Fig Tree and Onverwacht
groups. Longitudinal faults occur in the fold
flanks, mostly removing the anticlines; the belt
is mostly structured as a series of synclines juxtaposed by faults. One of the major faults, the
Inyoka–Saddleback fault system, runs along the
Northern edge of the Saddleback syncline and
separates two terranes within the belt, with
different stratigraphies and ages. This major
structure was interpreted as the main thrust of
the ca. 3.2 Ga orogeny (De Ronde and De Wit,
1994; De Ronde and Kamo, 2000; De Wit et al.,
1983; Lowe, 1999; Lowe et al., 1999), subse-
quently steepened by the D2b exhumation; the
Inyoni Shear Zone is now regarded as its lower
crustal expression.
Site description: The morphological role of the
Moodies group is evident in this landscape, as it
occupies the summits dominating the Fig Tree
valley, in front of us. The complexly folded
structure of the belt is hinted by the structural
repetitions of individual layers on the flank of
the valley.
Walking or driving 100–200 m down along the
road, it is possible to observe some crushed
sandstones (probably from the Fig Tree group),
corresponding to the trace of the Inyoka
fault.
Bibliography
Anhaeusser, C., Robb, L., and Viljoen, M. (1983). Notes on the provisionnal geological map of the
Barberton greenstone belt and surrounding granitic terrane, eastern Transvaal and Swaziland
(1:250 000 color map). Transactions of the Geological Society of South Africa, 9:221–223.
Anhaeusser, C. R. (1983). The geology of the Schapenburg greenstone remnant and surrounding
Archaean granitic terrane south of Badplaas, Eastern Trasvaal. In Contributions to the geology
of the Barberton Mountain land, volume 9 of Special publications, pages 31–44. Geological
Society of South Africa.
Belcher, R. W., Kisters, A. F., Poujol, M., and Stevens, G. (2005). Structural emplacement of
the 3.2 ga Nelshoogte pluton: implications for the origin of dome-and-keel structures in the
Barberton granite-greenstone terrain. In Geocongress, Durban.
De Ronde, C. and De Wit, M. (1994). The tectonic history of the Barberton greenstone belt,
South Africa: 450 million years of Archean crustal evolution. Tectonics, 13:983–1005.
De Ronde, C. and Kamo, S. (2000). An Archaean arc-arc collisional event: a short-lived (ca 3
Myr) episode, Weltvreden area, Barberton greenstone belt, South Africa. Journal of African
Earth Sciences, 30(2):219–248.
Day 4: Tuesday, 10 July
49
De Wit, M. J., Fripp, R., and Stanistreet, I. (1983). Tectonic and stratigraphic implications of
new field observations along the southern part of the Barberton greenstone belt. In Anhaeusser,
C. R., editor, Contributions to the Geology of the Barberton Mountain Land, volume 9 of Special
publications, pages 21–29. Geological Society of South Africa.
Faure, K. and Harris, C. (1991). Oxygen and Carbon Isotope Geochemistry of the 3.2 Ga
Kaap Valley Tonalite, Barberton Greenstone-Belt, South-Africa. Precambrian Research, 52(34):301–319.
Heubeck, C. and Lowe, D. R. (1994a). Depositional and Tectonic Setting of the Archean Moodies
Group, Barberton Greenstone-Belt, South-Africa. Precambrian Research, 68(3-4):257–290.
Heubeck, C. and Lowe, D. R. (1994b). Late Syndepositional Deformation and Detachment
Tectonics in the Barberton Greenstone-Belt, South-Africa. Tectonics, 13(6):1514–1536.
Kröner, A. and Compston, W. (1988). Ion microprobe ages of zircons from early Archaean granite
pebbles and greywacke, Barberton greenstone belt, Southern Africa. Precambrian Research,
38:367–380.
Lowe, D. R. (1999). Geological evolution of the Barberton greenstone belt and vicinity. Geological
Society of America Special Paper, 329:287–312.
Lowe, D. R., Byerly, G., and Heubeck, C. (1999). Structural divisions and development of the
west-central part of the Barberton Greenstone Belt. Geological Society of America Special
Paper, 329:37–82.
Robb, L., Barton, J.M., J., Kable, E., and Wallace, R. (1986). Geology, geochemistry and isotopic
characteristics of the Archaean Kaap Valley pluton, Barberton mountain land, South Africa.
50
Day 5: Wednesday, 11 July
The ca. 3.1 Ga GMS suite and associated deformation
S26◦ 05.007’; E30◦ 26.728’
Stop 5.1
Pavements of the ca. 3.1 Ga Heerenveen granite
Outcrops of the late-stage 3.1 Ga Heerenveen granite, containing prominent solid-state fabrics.
Access : Take the road opposite the Badplaas “Aventura” and head south, passed the Badplaas post
office, direction Chrissiesmeer. Drive for about 23 km (for most parts in granites related to the Heerenveen
batholith); stop on the dirt road and walk for ca. 100 m to your left (east) towards a pavement at the edge
of a pine tree plantation (total 23 km).
Aim: Introduce the Heerenveen batholith, the rock types forming it and the role of synmagmatic deformation.
Context: The Heerenveen batholith forms part
of the late-stage 3.1 Ga potassic suite of granite
plutons that largely conclude the main phases
of plutonism around the Barberton Mountain Land. These late-stage granites typically
form laterally extensive (up to several thousand squarekilometres) but thin (< 1km) subhorizontal sheets (Hunter, 1973; Anhaeusser
and Robb, 1983; Westraat et al., 2004). The
granites are typically heterogeneous ranging
in composition from granodiorites, monzogranites, syenogranites to, locally, syenites — the
“GMS suite” after Yearron (2003). They consist of quartz, K-feldspar (mainly microcline),
plagioclase, minor muscovite and biotite (in
places chloritized) with accessory amounts of
zircon. Multiple intrusive relationships are
common and most of the large batholiths appear to have been assembled through a multitude of granite sheets.
acterized by lit-par-lit intrusive relationships
between phases related to the Heerenveen granite and older TTG gneisses. The granites,
aplites and pegmatites occur either as isolated
sills and dykes or as sheeted sill-complexes.
At higher structural levels, the dykes and sills
form an anastomosing network separating large
angular rafts of TTG basement gneisses. At
the highest structural levels, exposed on top
of the western escarpment and in the center of
the Heerenveen batholith, the granite appears
rather homogeneous over large areas. Two main
textural phases can be distinguised including
(1) a medium-grained, locally porphyritic granite with K-feldspar megacrysts, and (2) a finegrained, homogeneous granite. These structurally higher phases locally contain rafts of
older TTG gneisses, that display variable degrees of assimilation, often represented by faint
ghost structures, or variably gneissose xenoliths
of earlier granite phases and/or aplite and pegThe Heerenveen batholith covers an area of ap- matite sills/dykes.
proximately 600 km2 . The composite granite
sheet is floored by banded TTG gneisses and Site description: Individual features to be
enclosed greenstone remnants while the roof seen:
rocks are not exposed.
• These particular pavements show a relAt least four main intrusive phases can be disatively homogeneous, medium-grained
tinguished. The granite is structurally zoned
phase of the Heerenveen batholith infrom bottom to top. The lower parts are chartruded by pegmatite veins.
51
52
Day 5 : Wednesday, 11 July
• Relatively
sporadic
K-feldspar
megacrysts, often magmatically zoned.
These megacrysts are common for the
potassic batholiths throughout the Barberton Mountain Land. They may occur
as clusters, making up to of 20-25 % of
the rock, as isolated megacrysts or may
be absent altogether, particularly in finegrained phases of the batholith.
• N(N)E-trending, steeply-dipping solidstate gneissosity. This gneissosity is developed almost throughout the Heerenveen batholith. It is regionally developed
and also prominent in adjacent batholiths
Stop 5.2
such as the large Mpuluzi batholith to
the immediate east. Notably, the NEtrending fabric appears largely confined
to rocks of the 3.1 Ga granite suite.
• Intrusive pegmatites. Pegmatite dykes
are a common occurrence in the potassic granites. Numerous generations of
pegmatites can be distinguished based on
cross-cutting relationships with different
granite phases as well as the deformation
of different dyke generations that may appear folded, boudinaged and, in places,
mylonitized. In this outcrop, pegmatites
are late-stage intrusives.
S26◦ 06.567’; E30◦ 28.367’
Intrusive breccias in the Heerenveen batholith
Access : Drive back on the main gravel road towards Badplaas for ca. 1.5 km and turn right into a forestry
road after ca. 1.5 km; continue with the forestry road for ca. 2 km. On the right hand side (SW) is a large
pavement (total 5 km).
Aim: Illustrate the multiple intrusive relationships and intrusive breccias in the 3.1 Ga
Heerenveen batholith.
Context: This outcrop forms part of an up to 2
km wide transition zone between sheeted granites confined to synmagmatic shear zones (next
outcrops) and the homogeneous, megacrystic
central granites of the Heerenveen batholith.
and solid-state fabric, which is intruded and
brecciated by a medium-grained leucogranite
that contains the same solid-state fabric as the
megacrystic granites. Up to six phases of brecciation can be recorded in some of the outcrops
surrounding this pavement, testifying to the
Site description: Two main phases can be
fact the the Heerenveen batholith was incredistinguished on this pavement — an earmentally assembled through a number of dislier megacrystic granite of the Heerenveen
tinct phases.
batholith showing a NE trending magmatic
53
S26◦ 10.017’; E30◦ 26.867’
Stop 5.3
Synmagmatic shear zones in the Heerenveen batholith
Access : Return to the main gravel road and turn left (SW). After ca. 4 km, turn to your left (SE) into a
smaller forestry road. Continue with this road for ca. 5 km until you cross some pavements in a small creek
(first stop). To reach the second stop — continue on foot to the north along the creek. After ca. 500 m and
a bend in the river, there are some large pavements to the left and right of the creek. These constitute the
second stop (S26◦ 09.767’; E30◦ 26.933’) (total 11 km).
Aim: Illustrate multiple, syntectonic granite sheeting and positive feed back between deformation (strain localization) and granite sheeting - synmagmatic shear zones
1410
R.W. Belcher, A.F.M. Kisters / Journal of Structural Geology 28 (2006) 1406e1421
Fig. 3. Geological map of the Heerenveen batholith showing: (a) The main intrusive phases and their distribution, reflecting the overall zonation of the batholith,
consisting of a central core of relatively homogeneous megacrystic granite, bound by compositionally heterogeneous marginal zones. (b) The four assembly stages
Figure
22:
Geological
structural
map
of the
Heerenveen
batholith
and Kisters,
2006).
(stage
1 e oldest,
to stage 4 eand
youngest)
based on mainly
cross-cutting
relationships,
but also fabric
development(Belcher
and internal geometry.
The correlation
between Left:
the compositionally
phases and
timing is detailedintrusion
in Table 1. (c) The
distributionRight:
of magmaticStructural
and solid-state fabrics
batholith and gneissosities
Lithological
map different
showing
thetheirsuccessive
phases.
mapin the
indicating
the trend of
in the surrounding basement. Note the magmatic foliation in the central, homogeneous megacrystic granite and the solid-state foliation and associated lineation in
foliation,
and the
position
thecentral
syn-magmatic
the surrounding
granites.
The marginof
of the
megacrystic granitesshear
is bound zones.
by two synmagmatic shear zones and corresponds to the zones of heterogeneous
granite sheeting.
extent of ca. 350 m showing a crude vertical, internal zonation.
The baseProtomylonitic
of the complex is characterized
by isolated
sheets
Context:
and mylonitic
fabrics
(1e2 cm and up to 2 m thick) that intrude parallel to the shaloccurlowly
in dipping
threegneissosity
distinct,
several
hundred
and compositional banding ofmethe
TTG gneisses
(Fig. 5a). the
This Heerenveen
basal zone can be
traced for
ter wide
belts within
batholith
20 km at approximately the same elevation along the
and over
along
its western margin. The high-strain
eastern margin of the Heerenveen batholith. At this structural
zoneslevel,
show
a sheets
compositional
and
intrusive
constitute betweenbanding
ca. 5 and 25%
of thea
outcrop. Most
of the sheets
are pegmatitic
with by
up to
10 cm
penetrative
gneissic
foliation
defined
quartz
large, euhedral K-feldspar crystals intergrown with quartz
ribbons
thatmuscovite.
alternate
withaplites
finely
recrystaland minor
Fine-grained
are also
common,
lizedwhile
quartz-feldspar
aggregates.
Microcline
leucogranitic sheets are
rare. At higher structural
levels,
within ca. 100
m from rounded,
the basal zone, granitic
sheets become
megacrysts
form
centimeter-sized
more abundant and may constitute 50e60% of the outcrop.
mantled
in coalescing
augenThe dm-porphyroclasts
to m-wide sheets formresulting
a branching and
network gneisses
of foliation-parallel
and cross-cutting
low-angle
textured
whilesillslarge
feldspars
from
sheets that engulf rafts of the TTG basement (Fig. 5b). Fineto medium-grained leucogranites increase in abundance, while
pegmatites become subordinate. At the highest structural
levels exposed, within ca. 250 m from the basal zone, granitic
sheets dominate and constitute >80% of the outcrop. Significantly,
isolated country-rock
screens between
the intrusivealong
pegmatites
are commonly
fragmented
granitoids retain their shallow SE dips with only little evidence
microfaults,
resulting
in
bookshelf-type
of rotation compared to country-rock gneisses outside the east-strucern
lit-par-lit complex.
tures.
The shallowly dipping granitoid sheets contain a well developed, sheet-parallel, high-temperature, solid-state foliation,
particularly
in the lower parts fabrics
of the lit-par-lit
de- for
The protomylonitic
can complex,
be traced
fined by the grain-shape preferred orientation of quartz and
several
kilometers
and
up
to
20
kilometers
quartzefeldspar aggregates and the orientation of phyllosilicates,
muscovite.
with the fabrics
gneissosityare
is a subalongmainly
strike.
TheAssociated
high-strain
down-dip
defined
by stretchedfoliation
quartz- and quartze
parallellineation
to the
magmatic
and solidfeldspar aggregates and muscovite. Both the solid-state foliastate
gneissosity
ofin the
Heerenveen
batholith
tion
and stretching
lineation
the intrusive
sheets are parallel
to
the aplanar
and linear
fabrics ofinthethe
older
country-rockof the
and
gradual
increase
intensity
gneisses (Figs. 3c and 6a, b). It is clear that, on a regional
foliation into the shear zones is recorded. In
scale, the country-rock gneisses have acquired their fabric during the main phase of tectonism in the granitoid-greenstone
terrain at ca. 3230 Ga (Dziggel et al., 2002; Stevens et al.,
2002). This suggests an almost coaxial overprint of these older
54
detail, the shear fabrics undulate, but define
ENE-WSW trends on a regional scale with subvertical to steep southerly dips. Stretching lineations defined by rodded quartz- and quartzfeldspar aggregates show shallow- to moderate
SW plunges, parallel to the lineation developed
across most of the Heerenveen batholith.
A notable exception from this general ENEtrend of high-strain zones are two northerly
trending segments along the western margin
of the Heerenveen batholith. The northerly
trending parts of the western boundary of the
Heerenveen batholith show subvertical, downdip stretching lineations. Shear sense indicators are common and include S-C fabrics,
together with σ-clasts and brittle microfaults
in tiled K-feldspar megacrysts. The ENEtrending shear zones consistently show a component of dextral strike-slip. Shear-sense indicators along the northerly trending shear zones,
in contrast, consistently show a component of
sinistral strike-slip. Notably, shear sense indicators are only observed in horizontal sections, perpendicular to the subvertical stretching lineations. The subvertical stretch combined with the strike-slip movement suggests
that the northerly trending shear zones are
sinistral transpressive shear zones. Provided
that the northerly trending transpressive shear
zones and the ENE trending, dextral strike-slip
shear zones formed at the same time, the highstrain zones within and along the western margin of the Heerenveen batholith form a conjugate shear zone pattern.
ing characterized by dyke-in-dyke intrusive relationships between a multitude of granitic,
pegmatitic, and aplitic phases. Individual
granitic sheets vary in width from < 1 m
to > 10 m showing steep, mainly southeasterly dips and ENE- to NE trends parallel to
the solid-state gneissosity and the shear-zone
boundaries. Intrusive relationships can still be
discerned and intrusive breccias are well preserved, particularly at lower fabric intensities.
Angular fragments of the megacrystic granite
contain magmatic and solid-state fabrics that
are parallel to the solid-state fabric in the intrusive fine-grained leucogranitic sheets. Notably,
the solid-state fabric intensity in many of the
late-stage intrusive sheets may be considerably
higher compared to the wall rocks that they intrude.
This strain localization into intrusive granitoids
together with the, on a regional scale, close
spatial relationship between high-strain fabrics
and granitic sheeting suggests synmagmatic deformation along the shear zones. The positive
feed back between the presence of melts and
deformation is also illustrated by the abrupt
along-strike termination of the two prominent,
ENE-trending shear zones close to the eastern
margin of the Heerenveen batholith. There is
no evidence of the several hundred meter wide
shear zones outside the Heerenveen batholith in
the older TTG country-rock gneisses.
Site description: Pavements in this area show
the protomylonitic nature and the multiple inThe occurrence of high-strain fabrics closely trusions associated with the synmagmatic shear
corresponds to zones of extensive granite sheet- zone.
55
ca. S26◦ 11.017’; E30◦ 32.117’
Stop 5.4
Schapenburg Greenstone remnant
Metamorphism and deformation in a greenstone remnant along the Southern extension of the ISZ
Access : Drive back to the main dirt road; turn left. Continue for 9 km until you arrive at the R33.
Turn left, direction Amsterdam/Mbabane. After 8.3 km, arrive at the N17/R33; turn left, direction Amsterdam/Mbabane. Pass the Jessievale sawmill; leave Lothair Road on right. 3.1 km after this road, turn left
on an unsignaled forestry road (you’re 0.5 km before the R33 turnoff towards Amsterdam). Drive 4.3 km on
this road through pine tree plantations, pass the farm on your left; continue for a further 1 km; keeping on
the left hand side of the valley. Stop at viewpoint, on Karoo dolerites (total 30 km).
Aim: (a) Show amphibolite-facies greenstone lithologies of Fig Tree and Onverwacht Group
age with well preserved sedimentary and volcanic textures; (b) Demonstrate the syn- to
post-tectonic nature of peak metamorphism and discuss timing of metamorphism and the
significance of the peak P-T conditions; (c) Discuss the extend and effect of the ca. 3.1
Ga deformation.
Figure 23: Geological map of the Schapenburg Greenstone remnant (Stevens et al., 2002).
Context: The Schapenburg schist belt is made
up of subvertical to steep southeasterly dipping
amphibolite-facies metasediments of the Fig
Tree Group. These rocks are overlain by mafic
and ultramafic rocks to the east, that can be
correlated with rocks of the Onverwacht Group
(Anhaeusser, 1983; Stevens et al., 2002). In the
east, the Schapenburg schist belt is bordered
by 3.1 Ga granites of the Mpuluzi batholith
and in the west, it is bordered by the Heeren-
veen batholith. The Theespruit River in the
north has eroded into banded TTG gneisses
and enclosed greenstone remnants, the largest
of which is the Schapenburg schist belt.
Site description:
5.4.a: Schapenburg schist belt, viewpoint.
This vista gives an overview of the extensive
granite-gneiss terrain to the south of the Barberton greenstone belt. The main valley be-
56
low us is occupied by the Schapenburg Shcist
Belt; the high grounds on either side correspond to the Heerenveen (left, or West) and
Mpuluzi (right, or East) batholiths (note the
prominent pavements that build up the escarpment in the east). The prominent hills in the
far north are part of the 3.1 Ga Boesmanskop
syenite and, on clear days, one might be able
to see the southernmost parts of the Barberton greenstone belt (parts of the Onverwacht
anticline and the Stolzburg syncline, the latter
developed in rocks of the Moodies Group).
chocolate-type, lying within or close-to the bedding/foliation of the greenstones. The orientation of folds and the chocolate-type boudinage indicate a bulk shortening strain at high
angles to NE-SW trending bedding of the
Schapenburg schist belt. The NW-SE shortening strain recorded in these localities corresponds to the regional NW-SE shortening direction during the emplacement of the subhorizontal granite sheets (Jackson and Robertson,
1983; Westraat et al., 2004). Some of these
pegmatite and granite veins are most likely related to older phases related to the Heerenveen
5.4.b (S26◦ 10.901’; E30◦ 32.240’): Fine-grained batholith, although monazite and zircon datphase of the Heerenveen batholith.
ing from pegmatites has illustrated that some
of these formed in association with the 3.23 Ga
Access: Ca. 200 m walk northwards from view- peak of metamorphism.
point towards granitic pavements.
Features to be seen: These pavements expose
a relatively fine-grained, homogeneous and regionally widespread variety of the Heerenveen
batholith. The granite appears largely undeformed, although a weak NE-trending foliation
and steep lineation can locally be seen. Intrusive relationships recorded further north indicate that the fine-grained homogeneous granite
is the last of the main, texturally distinct intrusive phases of the Heerenveen batholith. The
very weak solid-state fabric suggests that this
granite phase was emplaced during the waning
stages of regional NW-SE directed shortening.
A generation of late pegmatite dykes is intrusive into the Heerenveen batholith.
5.4.c (S26◦ 10.920’; E30◦ 32.241’): Contact between greenstones of the Schapenburg schist
belt and the Heerenveen granite.
Access: Walk due south and down the gentle
slope for ca. 50m, into a small gully.
5.4.d (S26◦ 10.858’; E30◦ 32.399’):
Crossbedded metasediments of the Fig-Tree Formation.
Access: Walk on the ridge on the right hand
bank of this gully, then follow the ridge down,
looking at the pavements on the way down.
(300 m) Visit outcrops around the two isolated
pine trees near this point, in a 25 m radius.
Features to be seen: Finely laminated volcanoclastics of Fig Tree Group affinity (Stevens
et al., 2002). Sedimentary features include
laminated beds and occasional cross-bedding.
Cross-beds consistently indicate that the volcanoclastics are younging towards the southeast, the direction in which they are overlain
by the Onverwacht Group mafic-ultramafic sequence. These rocks consist of the simple mineralogy, quartz + plagioclase feldspar + biotite
± K-feldspar. Bulk rock composition varies
from granodioritic to granitic and this, coupled
to the evidence for clastic processes has led to
the hypothesis that these rocks represent a distal felsic volcaniclastic tuff deposit. Apart from
the higher metamorphic grade, these rocks are
similar to other thick felsic to intermediate volcaniclastic units developed in the Fig Tree formation within the main body of the Barberton
greenstone belt.
Features to be seen: The knife-sharp contact
between the Heerenveen batholith and greenstones of the Schapenburg schist belt. Note the
absence of contact metamorphic effects, brecciation of wall-rocks nor any strain effects related to the granite emplacement on the wallrocks or the granites themselves. Note, however, on your way to the next outcrops (stop
5.4.d), there are numerous examples of peg- 5.4.e: Metaturbidites. Typical rhythmic bandmatites and granite-veins that undergo boud- ing (S26◦ 10.882’; E30◦ 32.529’), cordierite porinage and folding. Boudinage is mainly of a phyroblasts (S26◦ 10.940’; E30◦ 32.545’).
Access: Walk down the slope, due East from
the position of Stop 2d. Stop at the stream to
view the banded metaturbidites (200 m). The
cordierite porphyroblasts are best developed in
layers to the south and upwards along the slope
(50 to 100 m).
Features to be seen: The metasediments developed to the eastern of the mettatuff unit
consist of a 200 m thick sequence of interbanded mica + orthoamphibole ± garnet ±
cordierite schists and magnetite + orthoamphibole + quartz schists. Together, these
constitute a small scale (cm) and rhythmical
metapelite-metagreywacke banding. The more
metapelitic mica-rich bands are typically 10 cm
thick and locally, can be seen to grade into
more quartz-rich metagreywacke compositions
towards the eastern (top) side of individual layers. The metagreywacke layers are typically
4 to 5 cm thick. Collectively, these observations led Anhaeusser (1983) to the conclusion
that these banded rocks represent a metamorphosed turbidite sequence. The more aluminous and magnesian layers within this unit contain abundant cordierite porphyroblasts. These
are generally flattened in the plane of the foliation, and in some layers define a well-developed
down-dip lineation (averaging 143/76). In contrast, garnet and orthoamphibole, interpreted
to have grown later in the metamorphic history under peak metamorphic conditions, postdate the penetrative fabric defined by biotite,
quartz and cordierite. In hand specimen fanned
anthopyllite needles that overgrow the foliation are clearly visible. Mineral chemical zonation patterns indicate near complete equilibra-
57
tion under peak conditions of 640 ± 40◦ C and
4.8 ± 1.0 kbar (Stevens et al., 2002). The maximum age of metamorphism is defined by the
3231 Ma age of a syntectonic tonalite intrusion into the central portion of the schist belt,
while detrital zircons within the metasediments
have ages as young as 3240 Ma (Stevens et al.,
2002). Thus, sedimentation, burial to midcrustal depths, and amphibolite facies equilibration were achieved in a time span similar
to 15 Ma. The metasedimentary sequence is
separated from the overlying mafic-ultramafic
volcanic sequence by a band of quartzite that
Anhaeusser (1983) interpreted as a metamorphosed and recrystallised chert layer. In light of
the new age finding, this quartzitic layer most
likely represents an annealed structure along
which the Onverwacht and Figtree group rocks
were brought into juxtaposition prior to the
peak of metamorphism.
5.4.f (S26◦ 11.005’;
tonalite.
E30◦ 32.521’):
Intrusive
Access: Continue obliquely up the slope for 150
m, towards a small grey outcrop.
Features to be seen: This locality consists of a
small body of sheared tonalite emplaced close
to the contact between the Fig Tree group
metasediments and the overlying ultramafic
rocks of the Onverwacht group. This shear zone
is interpreted to be the site of intrusion of the
large syntectonic tonalite body that has been
dated and is characterized along much of its
exposed length by small tonalite lenses. These
carry a much stronger shear fabrics than the
main body.
58
Stop 5.5
S26◦ 05.765’; E30◦ 37.822’
Eagle Heights
Synmagmatic deformation and granite-sheeting during dextral transpression along the Welverdiend
shear zone
Access : Regain the N17 and turn left. Drive for 15 km and turn left (North) off the main road onto forestry
track following the sign “Eagle Height”. There’s an array of forestry roads, but try to keep left for most of the
time heading northwest for the main escarpment, some 9 km from the main road. The outcrops are around
the topo point 1732 m (total 29 km).
Aim: (a) Illustrate the multiple and structurally controlled granite emplacement of the
large, tabular ca. 3.1 Ga Mpuluzi batholith. (b) Multiple granite sheeting and dyking
during synmagmatic deformation and pronounced strain partitioning into intrusive dykes.
(c)Overview over southern granite-gneiss terrain and southern Barberton greenstone belt
and concluding remarks.
A S S E M B LY O F T H E M P U L U Z I BAT H O L I T H
Fig. 2. Geological map of t
gneiss terrane south of the B
greenstone belt illustrating t
distribution of the GMS sui
TTG gneisses and enclosed
remnants.
Figure 24: Geological map of the North-Western edge of the Mpuluzi batholith, showing the complex rock
assemblagegical
associated
the syn-magmatic
Welverdiend
zone (Westraat
2004)
of et
theal.,
GMS
suite in an extensional and rift-type t
results with
are presented
on older TTG
gneissesshear
and younger
was proposed by Kamo & Davis (1994), based o
potassic intrusive rocks to provide absolute age constraints on
nature of the rocks and the emplacement of some s
the timing of the emplacement and fabric development in differas NW–SE-trending, distinctly dyke-like bodies (F
ent igneous phases.
Context: The western margin of the Mpuluzi 600 m to the NE.
Hunter (1957, 1973) was probably the first to
batholith forms an up to 700m high escarpsubhorizontal,
sheet-likesolidgeometry of the Mpuluz
The
steeply
inclined,
NE–SW
trending
GMS suite
in the study
ment. TheThe
steep-sided
pavements
thatarea
underlie
also estimated
a thickness
of the granitoid shee
state gneissosity displayed
by all intrusive
rocks
the Northwestern
parts
of escarpment
visited
1000 m based on his mapping of the Archaean gr
Rocks of the
Mesoarchaean
GMS suite
in the area studied here
along the western escarpment is related to the
mountaineous terrain of Swaziland. The tabular
include three mainmade
igneous
namely the Mpuluzi batholith
here are predominantly
upunits,
of gneissose
of and
a synmagmatic
zone,
the
since been shear
confirmed
in regional
field studies b
(sensu
and the smaller Syenogranite
intrusions of the presence
Boesmanskop
rocks related
to lato)
the Boesmanskop
Welverdiend
shear
zone
that
records
dextral
(1980)
and
Anhaeusser
&
Robb
(1983),
who als
Weergevonden
syenogranites
situated
along
the
NW
margin
of
Complex, the type-locality of which is repretranspressive
very shallow crustal level of emplacement for
the Mpuluzi batholith (Fig. 2). The Mpuluzi
batholith is motion.
a
sented by the two prominent hills to the immebatholith mainly on the grounds of textural ev
composite pluton, made up of a number of petrographically and
diate North
of thedistinct
escarpment.
Leucogranites
Site description:
outcops
granitoids.
Theillustrate
lower- tothe
sub-greenschist-facies
texturally
phases that
range in composition
from grano- These
related todiorite,
the Mpuluzi
batholith
stricto) synkinematic
of granodioritic
dykesgreenstone belt to
conditions
of the Barberton
monzonite
and(sensu
monzogranite
to syenograniteinjection
are exposed
along the
someet 500–
during
deformation.
north render such shallow emplacement levels lik
(Anhaeusser
& escarpment,
Robb 1983; Robb
al. 1983;
Yearron
2003).
there are, as yet, no direct and reliable P–T d
The semicircular granitoid covers an area of at least 4000 km2
constrain the emplacement depth.
south of the Barberton greenstone belt (Fig. 1). It occupies the
The Mpuluzi batholith intrudes into older, c
high-lying peneplain between South Africa and Swaziland, and
amphibolite-facies, steeply dipping, banded TTG
borders against the low-lying, older TTG–greenstone terrane in
enclosed supracrustal greenstone remnants. Base
the north along a prominent, 500–700 m high escarpment. Its
are parallel to the western, strongly gneissose
southwestern extent is concealed by younger Karoo-aged cover
Mpuluzi batholith (Figs 2 and 3) and structural e
rocks. Other large batholiths of the GMS suite in the region
to the rotation of the wall-rock gneissosities into p
include the Nelspruit batholith to the north of the Barberton
this western margin (see below). Notably, a s
greenstone belt and the Heerenveen and Piggs Peak batholith in
subvertical, NE–SW-trending gneisses within and
the south and SE of the greenstone belt, respectively (Anhaeusser
Mpuluzi batholith has been described by Jackson
et al. 1981) (Fig. 1).
(1983) some 30 km SE of the present study area
U–Pb age constraints from zircons from a fine-grained
the granites of the Mpuluzi batholith have intruded
granodioritic phase indicate an age of crystallization of
most parts of the Barberton greenstone belt, the
3105 3 Ma for the Mpuluzi granite, whereas the main, coarse-
59
Figure 25: Conceptualized map of the outcrops around locality 5.5 (Sonke, 2006).
The dykes consist of K-feldspar augen that
commonly define a pervasive L or LS fabric. The medium-grey matrix consists of fine
grained K-feldspar, plagioclase, biotite and
quartz. The dykes brecciate gneisses related
to the Boesmanskop syenite, pointing to highstrain rates during dyke/sheet propagation.
Significantly, the intrusive dykes show considerably higher strain intensities compared the
wall-rock gneisses. This probably reflects a
strain partitioning into the magma-filled dykes.
Magmatic flow-fabrics, defined by aligned Kfeldspar laths, wrap around wall-rock xenoliths. The foliation dips steeply to the SE and
trends NE, parallel to the overall strike of the
Welverdiend shear zone, and lineation plunge
shallowly (15-25◦ ) to the NE, also parallel to
solid-state stretching lineations in this part of
the shear zone. Deformation continued after
full crystallization of the dykes and sub-solidus,
solid-state fabrics overprint magmatic fabrics.
Similar, closely spaced dykes can be observed
in a over 1 km wide, NE-SW trending corridor,
parallel to the Welverdiend shear zone along
the western escarpment.
Various generations of pegmatite dykes and
sheets can be observed. Intrusive relation-
ships between pegmatite dykes and the highlystrained granodioritic dykes point to their
largely contemporaneous timing. Note that
pegmatite dykes outside the granodioritic dykes
are tyically weakly deformed or undeformed,
parallel-sided sheets. Pegmatite dykes typically
show one of the following features, where they
intersect granodioritic dykes:
1. Pegmatite dykes are truncated by granodiorite dykes - in this case, the pegmatites are clearly older.
2. Pegmatites terminate against granodioritic dykes, but (a) pegmatites bulge
along the intersection, often showing a
quartz-rich core, and (b) pegmatite dykes
may form thin offshoots that protrude
into the granodioritic dyke or pegmtites
taper within the granodiorite. These offshoots and tapered terminations are commonly progressively transposed into the
fabric of the granodioritic dykes. In this
case, pegmatite intrusion post-dates that
of the granodioritic dykes, but the propagation of the pegmatite dykes was hampered by the presence of probably a melt
phase in the granodiorites.
60
Day 5: Wednesday, 11 July
3. There are numerous localities where the
intersection between pegmatite dykes and
granodiorites are characterized by an embayment of the latter into the former. We
interpret this to represent a “collapse” of
the melt-filled shear zone (the granodioritic dykes) into the fracture related to
pegmatite propagation.
4. Pegmatites cross-cut the granodiorites,
but are deflected into parallelism or at low
angle with the dyke margins. Pegmatites
undergo various degree of folding, boudinage and complete dismemberment. In
this case, pegmatite emplacement is likely
to have occurred when the granodiorites
were, for most parts, fully crystallized.
Deformation occurred mainly in the solidstate.
5. Pegmatites dykes cross-cut granodioritic
dykes without evidence of deformation.
This latest generation of pegmatites
dykes is post-kinematic.
Bibliography
Anhaeusser, C. R. (1983). The geology of the Schapenburg greenstone remnant and surrounding
Archaean granitic terrane south of Badplaas, Eastern Trasvaal. In Contributions to the geology
of the Barberton Mountain land, volume 9 of Special publications, pages 31–44. Geological
Society of South Africa.
Anhaeusser, C. R. and Robb, L. J. (1983). Geological and geochemical characteristics of the
Heeenveen and Mpuluzi batholiths south of the Barberton greenstone belt, and premliminary
thoughts on their petrogenesis. In Contributions to the geology of the Barberton Mountain
land, volume 9 of Special publications, pages 131–151. Geological Society of South Africa.
Belcher, R. W. and Kisters, A. F. (2006). Progressive adjustments of ascent and emplacement
controls during incremental construction of the 3.1 Ga Heerenveen batholith, South Africa.
Journal of Structural Geology, 28:1406–1421.
Hunter, D. (1973). The granitic rocks of the precambrian in Swaziland. In Contributions to
the geology of the Barberton Mountain land, Special publications, pages 131–145. Geological
society of South Africa.
Jackson, M. and Robertson, D. (1983). Regional implications of Early Precambrian strains in the
Onverwacht Group adjacent to the Lochiel granite, North-West Swaziland. In Contributions
to the geology of the Barberton Mountain land, volume 9 of Special publications, pages 45–62.
Geological Society of south Africa.
Sonke, G.-J. (2006). Internal architecture of the sheeted margin of the 3.1 Ga Mpuluzi batholith,
Barberton granite-greenstone terrain. Honors thesis, Stellenbosch University.
16.
Yearron, L. (2003). Archaean granite petrogenesis and implications for the evolution of the
Barberton mountain land, South Africa. Unpub. phd thesis, Kingston University.
Day 6: Thursday, 12 July
Return to Johannesburg and Capetown
Drive back to Johannesburg airport via Machadodorp, the N4 to Middleburg and the N12.
(ca. 315 km, 3 hours plus stops).
Participants returning to Cape Town fly out on flight 1 Time IT 109 at 12h50. Arrive Cape
Town at 15h00.
61
Part III
Articles
Day 6 : Thursday, 12 July
65
ecent articles (2003–2007) from the Stellenbosch University group have been included, as a
way to give the interested reader more details on the geology of the Barberton GraniteGreenstone terrain and on our models. The papers are arranged by theme, corresponding to
out three main research directions: the ca. 3.2 Ga accretionary orogen; the nature and origin of
TTG magmas; the emplacement of the ca. 3.1 Ga GMS suite.
R
The 3.2 Ga orogeny: structures and metamorphism
1. Kisters, A. F., Stevens, G., Dziggel, A., and Armstrong, R. (2003). Extensional detachment
faulting and core-complex formation in the southern Barberton granite–greenstone terrain,
South Africa: evidence for a 3.2 Ga orogenic collapse.Precambrian Research, 127:355–378.
2. Stevens, G. and Moyen, J.-F. (in press). High-pressure, low-temperature metamorphism
in the Barberton greenstone belt; a key to understanding Archaean tectonic evolution.
In Van Kranendonk, M., Smithies, R. H., and Bennet, V., editors, Earth’s oldest rocks,
Developments in Precambrian Geology, Chapter 5.7. Elsevier. (Uncorrected proofs)
3. Moyen, J.-F., Stevens, G., and Kisters, A. F. (2006). Record of mid-Archaean subduction
from metamorphism in the Barberton terrain, South Africa. Nature, 443:559–562.
4. Diener, J., Stevens, G., Kisters, A. F., and Poujol, M. (2005). Metamorphism and exhumation of the basal parts of the Barberton greenstone belt, South Africa: Constraining the
rates of mid-Archaean tectonism. Precambrian Research, 143:87–112.
Petrology and geochemistry of the TTG suite
5. Clemens, J.D., Yearron, L.M., and Stevens, G. (2006). Barberton (South Africa) TTG
magmas: geochemical and experimental constraints on source-rock petrology, pressure of
formation and tectonic setting. Precambrian Research, 151:53–78.
6. Moyen, J.-F., Stevens, G., Kisters, A. F., and Belcher, R. W. (in press). TTG plutons of the
Barberton granitoid-greenstone terrain, South Africa. In Van Kranendonk, M., Smithies,
R. H., and Bennet, V., editors, Earth’s Oldest rocks, Developments in Precambrian geology,
Chapter 5.6. Elsevier. (Uncorrected proofs)
Emplacement of the ca. 3.1 Ga GMS suite
7. Belcher, R. W. and Kisters, A. F. (2006a). Progressive adjustments of ascent and emplacement controls during incremental construction of the 3.1 Ga Heerenveen batholith, South
Africa. Journal of Structural Geology, 28:1406–1421.
8. Westraat, J. D., Kisters, A. F., Poujol, M., and Stevens, G. (2004). Transcurrent shearing,
granite sheeting and the incremental construction of the tabular 3.1 Ga Mpuluzi batholith,
Barberton granite-greenstone terrane, South Africa. Journal of the Geological Society of
London, 161:1–16.
Precambrian Research 127 (2003) 355–378
Alexander F.M. Kisters a,∗ , Gary Stevens a ,
Annika Dziggel b,1 , Richard A. Armstrong c
Extensional detachment faulting and core-complex formation
in the southern Barberton granite–greenstone
terrain, South Africa: evidence for a
3.2 Ga orogenic collapse
b
Received 6 February 2003; accepted 1 August 2003
a Department of Geology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa
Department of Geology, Economic Geology Research Institute, University of the Witwatersrand, Johannesburg, South Africa
c Research School of Earth Sciences, The Australian National University, Canberra, 0200 ACT, Australia
Abstract
The Barberton greenstone belt in South Africa is an Early- to Mid-Archaean, very low-grade metamorphic supracrustal belt
that is bordered in the south by a mid- to lower crustal gneiss terrain. Detailed mapping of the contacts between the supracrustal
and gneiss domains along the southern margin of the greenstone belt shows that the supracrustal rocks are separated from the
high-grade metamorphic gneiss terrain by an extensional detachment that is situated at and close to the base of the belt. The
extensional detachment is approximately 1-km wide and its location corresponds with the heterogeneous, mélange-like rocks
of the Theespruit Formation. Within the detachment zone, two main strain regimes can be distinguished. Amphibolite-facies
rocks at and below the granite–greenstone contacts are characterized by rodded gneisses and strongly lineated amphibolite-facies
mylonites. The bulk constrictional deformation at these lower structural levels records, in a subhorizontal orientation, the vertical
shortening and horizontal, NE–SW directed stretching of the mid-crustal rocks. The prolate, coaxial fabrics are overprinted by
greenschist-facies mylonites at higher structural levels that cut progressively deeper into the underlying high-grade basement
rocks. These mylonites have developed during non-coaxial strain and kinematic indicators consistently point to a top-to-the-NE
sense of movement of the greenstone sequence with respect to the lower structural levels. This relationship between bulk
coaxial NE–SW stretching of mid-crustal basement rocks and non-coaxial, top-to-the-NE shearing along retrograde mylonites
at upper crustal levels is consistent with an extensional orogenic collapse of the belt and the concomittant exhumation of
deeper crustal levels. The exhumation was initiated under amphibolite-facies conditions at depths of approximately 18 km. The
extensional collapse is coeval with or shortly follows the main D2 collisional event in the Barberton greenstone belt at ca.
3230–3220 Ma. Voluminous plutonism at ca. 3225 Ma along the northern margin of the belt is possibly related to the orogenic
collapse and associated decompression melting of lower crustal rocks. The extensional collapse coincides with the onset of the
coarse-clastic Moodies Group sedimentation which suggests that the small, isolated Moodies basins formed as supradetachment
basins in the collapsing hanging wall of the detachment. The steepening of lithologies and fabrics to their present-day vertical
∗ Corresponding author. Tel.: +27-21-8083113; fax: +27-21-8083129.
E-mail address: [email protected] (A.F.M. Kisters).
1 Present address: Institute of Mineralogy and Economic Geology, Aachen University of Technology (RWTH), Wuellnerstr. 2, 52056 Aachen,
Germany.
0301-9268/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.precamres.2003.08.002
356
A.F.M. Kisters et al. / Precambrian Research 127 (2003) 355–378
attitudes is ascribed to a late-stage solid-state diapiric component of the exhumed hot and buoyant basement gneisses that underlie
the relatively cool and dense mafic and ultramafic supracrustal succession.
© 2003 Elsevier B.V. All rights reserved.
formed during the diapiric ascent of the intruding
granitoids (e.g. Anhaeusser, 1984, 2001). In these
models, the denser greenstones are thought to be
infolded between rising diapirs and tilted to their
present-day subvertical attitudes. Related processes
of gravity-driven, but solid-state vertical tectonics
have recently been suggested by a number of studies to account for the dome-and-keel provinces of
many Archaean and Palaeoproterozoic terranes (e.g.
Marshak et al., 1997; Collins et al., 1998). Subsequent studies established that the BGB represents a
polyphase-deformed fold-and-thrust belt that formed
during two main accretionary events, D1 and D2 , at
ca. 3445 and 3225 Ma. These collisional events were
associated with episodes of voluminous calc-alkaline
tonalite-trondhjemite-granodiorite (TTG) plutonism
probably in an arc-trench environment (de Wit et al.,
1987, 1992; Armstrong et al., 1990; de Ronde and de
Wit, 1994; Kamo and Davis, 1994; Lowe, 1994, 1999;
Kröner et al., 1996). Despite the spatial and temporal
relationship between the TTGs and greenstones, most
of these studies tended to regard the structural evolution of the greenstone belt essentially in isolation and
focused on the central parts of the BGB where the
rocks have been affected by only lower grades of metamorphism and where strain intensities are relatively
low. Consequently, most current models invoke mainly
thin-skinned tectonic processes and the TTG plutons
are either considered to be passive and rigid basement
blocks onto which the supracrustals were thrusted, or
syntectonic igneous complexes that were magmatically accreted at the base of the thrusted greenstone
sequence (e.g. Williams and Furnell, 1979; Fripp
et al., 1980; de Wit, 1983; de Wit et al., 1983, 1987;
de Ronde and de Wit, 1994; Lowe and Byerly, 1999).
Recent studies on the metamorphic history of the
ca. 3445 Ma TTG terrain to the immediate south of
the BGB (Dziggel et al., 2002; Stevens et al., 2002)
hint at a very different tectonic evolution of the Barberton granite–greenstone terrain. These studies show
that granitoids and enclosed metasedimentary rem-
Keywords: Barberton greenstone belt; Archaean tectonics; Orogenic collapse; Extensional exhumation
1. Introduction
The Barberton granite–greenstone terrain in South
Africa is one of the world’s best preserved and most
extensively studied Early- to Mid-Archaean crustal
segments that has served as a type locality for our understanding of early crustal evolution (e.g. Viljoen and
Viljoen, 1969; Anhaeusser, 1969, 1984; de Wit et al.,
1983, 1987; Armstrong et al., 1990; Kröner et al.,
1991; de Ronde and de Wit, 1994; Kamo and Davis,
1994; Lowe, 1994; Lowe et al., 1999; de Ronde and
Kamo, 2000; amongst many others). Like many other
Archaean granite–greenstone terrains, it consists of
two main components: a polyphase-deformed, mainly
low-grade metamorphic supracrustal belt, the Barberton greenstone belt (BGB), and a surrounding complex granitoid-gneiss terrain (e.g. Anhaeusser, 1969;
Anhaeusser et al., 1981; Robb and Anhaeusser, 1983;
de Ronde and de Wit, 1994; Lowe and Byerly, 1999)
(Fig. 1). The granite-greenstones describe, in a regional context, a dome-and-keel geometry typical of
many Archaean provinces. Contacts between granitoids and greenstones in the BGB are commonly characterized by higher metamorphic grades compared to
that of rocks in the main belt, and although intrusive
relationships are locally preserved, most contacts are
tectonized with evidence of a complex deformation
history (Fripp et al., 1980; Anhaeusser, 1984; de Wit
et al., 1983, 1987; Kisters and Anhaeusser, 1995).
The dome-and-keel geometry and granite–greenstone contact relationships have been interpreted by
two entirely different evolutionary models that are
at the heart of the central and ongoing controversy
about Archaean tectonic processes, namely the role
of vertical versus horizontal tectonics for the formation and early reworking of the earliest continental
nuclei (e.g. Hamilton, 1998; de Wit, 1998; Marshak,
1999). Early tectonic models for the BGB interpreted granite–greenstone contact relationships as
intrusive contacts, containing contact metamorphic
aureoles and migmatites that were progressively de-
357
path that involved some 20–30 km of differential
uplift between the high-grade TTG terrain and the
low-grade rocks of the BGB. Moreover, equilibration
of the peak-metamorphic assemblages occurred at ca.
3230 Ma (Dziggel et al., 2002; Stevens et al., 2002),
i.e. some 200 Ma later than the intrusion of the south-
Fig. 1. Schematic geological map of the southern parts of the Barberton granite–greenstone terrain and its location in South Africa (inset)
(modified after Anhaeusser et al., 1981; de Ronde and de Wit, 1994). The Barberton greenstone belt is bounded in the south by ca.
3445 Ma trondhjemitic gneisses of the Stolzburg, Theespruit and Doornhoek plutons and in the northwest by ca. 3.2 Ga tonalites and
trondhjemites of the Kaap Valley tonalite and Nelshoogte pluton (ages after Kamo and Davis, 1994; de Ronde and Kamo, 2000). The ca.
3216 Ma Dalmein pluton sharply truncates structures of the Barberton greenstone belt and indicates a post-tectonic emplacement. Younger,
ca. 3.1 Ga granitoids of, e.g. the Mpuluzi and Heerenveen batholiths in the south form large, subhorizontal sheets. The extent of the study
area of the Stolzburg schist belt (Fig. 2) is indicated by the box.
nants contain peak-metamorphic assemblages that
record pressures of up to 11 kbar and temperatures of
ca. 700 ◦ C (Dziggel et al., 2002). The high pressure
estimates not only point to the burial of the granitoids and supracrustal remnants to mid- to lower
crustal levels, but also to a subsequent exhumation
358
The lithostratigraphic sequence of the BGB has traditionally been subdivided into three main groups on
the basis of dominant rock types (e.g. Visser, 1956;
Viljoen and Viljoen, 1969; Anhaeusser, 1969). From
the base upwards, the ca. 3.500–3.300 Ma Onverwacht
Group, made up of predominantly ultramafic- to
mafic volcanics, is overlain by argillaceous to arenaceous sediments and subordinate pyroclastics of the
2. Regional geology
3500–3450 Ma granite-gneiss terrain has long been
controversially discussed due to its structural, lithological and geochronological complexity (e.g. Viljoen
and Viljoen, 1969; de Wit et al., 1983, 1987;
Armstrong et al., 1990; Lowe et al., 1999). However,
to date no detailed kinematic or metamorphic studies
have been undertaken in this area, despite its obvious
significance for our understanding of the amalgamation of the granite–greenstone terrain.
ern TTG suite, but coinciding with the main period
of D2 collisional tectonics in the greenstone belt (de
Ronde and de Wit, 1994; Kamo and Davis, 1994).
Clearly, none of the existing tectonic models for the
BGB acknowledges the recycling, both the burial and
exhumation, of the southern TTG terrain into the deep
crust through potentially several tectonic cycles. The
juxtaposition of high-grade granite gneisses against
low-grade supracrustals, a common feature of numerous Archaean granite–greenstone terrains (e.g. Ridley
et al., 1997), raises questions as to the nature, location, geometry and timing of the structures that have
effectively decoupled the TTG terrain from the greenstone belt as well as the effects of these evidently
lithospheric-scale tectonic processes for the evolution
of the shallow-crustal BGB.
In this paper, we present the results of a study
into the structural evolution of granite–greenstone
contacts in the Stolzburg schist belt (SSB) in the
southwesternmost parts of the BGB (Figs. 1 and 2).
The southern contact of the BGB with the adjoining
Fig. 2. Simplified geological map of the Stolzburg schist belt (note that the map is broken up into two parts for better representation).
The narrow, E–W trending, subvertical supracrustal belt is bordered by mainly gneissic trondhjemites of the 3445 Ma Stolzburg pluton in
the south and the ca. 3236 Ma Nelshoogte pluton in the north. All lithologies and gneissosities in the adjoining gneisses are subvertical.
359
The Stolzburg schist belt forms the southwestern
extremity of the BGB (Figs. 1 and 2). The E–W trending belt is made up of subvertical, strongly schistose
meta-volcanosedimentary rocks that include mafic and
ultramafic volcanic rocks, felsic- to intermediate pyroclastics and volcanoclastics and minor chemical and
clastic sediments. These lithologies are correlated with
formations of the lower Onverwacht Group in the
south-central part of the BGB (Anhaeusser et al., 1981;
de Wit et al., 1983). The SSB is, on average, 1.5 km
wide and can be followed along strike for approximately 12 km. The belt widens towards the east where
it grades into relatively low-strain and low-grade metamorphic units of the main body of the BGB. The
greenstone sequence is bounded in the north by trondhjemitic gneisses of the 3236 ± 1 Ma Nelshoogte pluton (de Ronde and Kamo, 2000) and in the south by
variably deformed trondhjemitic gneisses of the ca.
3. Geology of the Stolzburg schist belt
thrusts and nappes during the NW-directed D2 tectonism (de Wit et al., 1987; de Ronde and de Wit, 1994).
Throughout this longlived and complex evolution,
the central parts of the BGB have only been affected
by lower- or sub-greenschist-facies metamorphism
(Xie et al., 1997). Upper-greenschist metamorphism
in the southern part of the belt has been interpreted
to be the result of an early seafloor and subsequent
burial metamorphism (Cloete, 1991) which is proposed to correlate with 3450–3490 Ma 40 Ar/39 Ar
ages found in komatiites of the lower formations of
the Onverwacht Group (Lopez-Martinez et al., 1984).
Narrow zones of amphibolite-facies rocks around
the margin of the BGB as they are developed in the
study area, are interpreted to represent either contact
metamorphic effects of the surrounding granitoids
(Anhaeusser, 1969; Cloete, 1991) or, in allochthonous
models for the BGB, as the metamorphic soles at
the base of the thrusted greenstone sequence (Fripp
et al., 1980; de Wit et al., 1987). In either case, considering the high-grade metamorphic nature of the
southern granite-gneiss terrain (Dziggel et al., 2002),
a very significant metamorphic break occurs across
the southern granite–greenstone contact that cannot
be reconciled with previous models for the evolution
of these contact zones.
ca. 3260–3225 Ma Fig Tree Group, which, in turn, is
overlain by the ca. 3.225–3215 Ma Moodies Group
that consists of mainly coarse-clastic sediments. The
early views of a continuous layer cake stratigraphy
were revised in subsequent studies that identified
significant structural repetitions in the sequence and
distinguished tectonostratigraphic domains within the
belt (e.g. Williams and Furnell, 1979; de Wit, 1983;
de Wit et al., 1983, 1987, 1992; Lamb, 1984;
Armstrong et al., 1990; de Ronde and de Wit, 1994;
Kamo and Davis, 1994; Kröner et al., 1996; Lowe
et al., 1999; de Ronde and Kamo, 2000). The correlation of deformational events between individual
domains is still controversial, but there is general
consensus that the belt was formed during two main
accretionary phases, D1 and D2 , at ca. 3445 and
3230 Ma that were both temporally associated with
episodes of voluminous TTG plutonism.
An early phase of subhorizontal thrusting (D1 ) and
recumbent folding in the Onverwacht Group in the
southern part of the BGB (de Wit et al., 1983, 1987)
is regarded to be responsible for the imbrication of the
lower mafic and ultramafic units of the Onverwacht
Group, the Komati Formation, with the underlying
Theespruit Formation along the Komati Fault (e.g.
Armstrong et al., 1990). The emplacement of the ca.
3445 Ma trondhjemitic plutons along the southern
margin of the BGB is thought to be synkinematic
with the D1 event (e.g. de Wit et al., 1987; de Ronde
and de Wit, 1994). The main phase of deformation,
D2 , occurred at ca. 3225 Ma. The D2 deformation
represents a NW–SE directed collisional event during which much of the present-day upright NE–SW
trending structural grain of the BGB, including folds
and thrust zones, was formed. It is believed to mark
the amalgamation of a northern and a southern terrane along the Saddleback-Inyoka Fault, the main
NE–SW trending terrane-bounding suture in the BGB
(de Ronde and de Wit, 1994; de Ronde and Kamo,
2000). The D2 event was associated with the intrusion
of the Kaap Valley and Nelshoogte plutons along the
northwestern margin of the belt (Fig. 1) and a change
in sedimentary style from Fig Tree Group sedimentation to the coarse-clastic Moodies Group deposition
(e.g. Lowe and Byerly, 1999; de Ronde and Kamo,
2000). The steepening of fabrics in the BGB to their
present-day upright attitudes is commonly attributed
to the progressive bulk shortening of, e.g. low-angle
360
The Stolzburg pluton is a medium- to coarse-grained
and compositionally relatively homogeneous trondhjemite. U–Pb zircon ages indicate an intrusion of the
pluton at ca. 3445 Ma that is, within error, similar to
zircon ages from the adjoining Theespruit and Doornhoek plutons to the east (Kamo and Davis, 1994)
(Fig. 1). Younger zircon and Pb/Pb titanite ages were
interpreted by Kamo and Davis (1994) to reflect a
later thermal event at ca. 3230 Ma.
The Stolzburg pluton is generally only weakly
foliated and appears massive in outcrop. In vertical
sections, however, the trondhjemites can be seen to
contain a prominent mineral lineation defined by
stretched plagioclase and quartz–plagioclase min-
3.1. Gneisses of the Stolzburg pluton
the SSB consists of greenschist-facies rocks and shows
lower strain intensities. The granite–greenstone contact relationships are particularly well exposed along
the southern contact of the SSB with the Stolzburg
pluton (Fig. 2). This area forms the focus of this study.
3445 Ma Stolzburg pluton (Kamo and Davis, 1994)
(Fig. 2). The western termination of the schist belt is
marked by the NW–SE trending, dyke-shaped intrusive of the 3107 Ma Kees Zyn Doorns syenite (Kamo
and Davis, 1994) and a northerly-trending, heterogeneous zone of complex schollen-and-raft migmatites
that is tentatively correlated with the Badplaas pluton to the west (Anhaeusser et al., 1981). The abrupt
deflection of the E–W trending greenstones of the
Stolzburg schist belt to N–S trends along the western termination of the belt is related to a late-stage,
NW–SE trending brittle–ductile fault.
The narrow, E–W trending SSB displays a pronounced axial symmetry. The northern and southern
margins of the belt are both marked by mylonitic
shear zones that may be several hundred metres
wide, developed at or close to the granite–greenstone
interface. The composite marginal zones consist
of amphibolite-facies and partially retrogressed
amphibolite-facies rocks that are interleaved with and
progressively replaced by greenschist-facies rocks towards the center of the belt. The central, axial zone of
Fig. 3. Outline of the Stolzburg schist belt showing L1 lineation domains in the supracrustal belt and the surrounding gneisses. All
stereographic projections are equal area projections and to the lower hemisphere. (a) Regionally developed, steep SE-plunging L1 lineations
in lineated trondhjemites of the Stolzburg pluton. (b) Steep E- to SE-plunging L1 stretching lineation in supracrustals in the southern parts
of the Stolzburg schist belt. (c) Poles to the gneissosity (dots) and stretching lineation (crosses) in gneisses of the Nelshoogte pluton. (d)
W- to SW-plunging L1 stretching lineations in supracrustals in the northern half of the Stolzburg schist belt.
361
Within ca. 200 m of the granite–greenstone contact, primary bedding (S0 ) in the amphibolite-facies
supracrustals of the SSB is only locally recognized
in relatively low-strain domains by, e.g. grain-size
variations in felsic volcanoclastics. For most parts,
S0 has been transposed into a subvertical, E–W
3.2. Fabric development in greenstones of the SSB
regime during fabric development (Fig. 5a). The mineral rods plunge at moderate- to steep angles to the
ESE and SE, parallel to the regionally developed L1
stretching lineation in the rest of the Stolzburg pluton (Figs. 3a and 4a). In the E, the Stolzburg pluton
appears largely undeformed, even where the plutonic
rocks are in proximity to the greenstone sequence
of the SSB. Cross-cutting relationships between the
Stolzburg pluton and the supracrustal greenstones testify to the primary intrusive contacts (Fig. 4) and large,
randomly orientated greenstone xenoliths that measure up to several tens of metres in size are relatively
common close to the granite–greenstone contact.
Fig. 4. Structural formline map of the S0 /S1 fabric in the Stolzburg schist belt. All stereographic projections are equal area projections and
to the lower hemisphere. (a) Poles to the S1 gneissosity and L1 lineations in trondhjemitic gneisses in the western parts of the Stolzburg
pluton. (b) Poles to the S0 /S1 fabric in supracrustals of the SSB. The fabric is predominantly subvertical and trends E–W. Note the easterly
dipping foliations and the overall great-circle distribution of S0 /S1 , outlining the easterly plunging F2 folding of the S0 /S1 fabric in the SSB.
(c) Shallow- to moderately E-pluning F2 fold axes (crosses) and poles to the S2 axial planar foliation (open circles). (d and e) Sketches of
mesoscopic F2 folds showing a characteristic Z-shape asymmetry along the southern margin of the SSB and S-shape asymmetry in the north.
eral aggregates. This lineation (L1 ) is present almost
throughout the Stolzburg pluton and plunges consistently at steep- to moderately angles to the E and
SE (Fig. 3a). Strongly gneissose fabrics only occur
within 300–600m of the granite–greenstone contact
(Fig. 4a) and a gneissic foliation (S1 ) is particularly
prominent in the westernmost parts of the pluton. S1
a high-temperature solid-state gneissosity defined by
quartz-ribbons and biotite foliae that wrap around
recrystallized plagioclase aggregates. The S1 gneissosity is parallel to the granite–greenstone contact
and dips at steep angles to the north (Fig. 4a). It
contains a moderate- to steep SE plunging mineral
stretching lineation parallel to the (L1 ) lineation developed in the lower strained central parts of the pluton (Figs. 3a and 4a). For most parts, gneisses of the
Stolzburg pluton are developed as strongly lineated
L > S tectonites. The rodding (L S) of all mineral
components is very pronounced within ca. 200 m of
the granite–greenstone contact and plagioclase and
quartz–plagioclase rods show aspect ratios of up to
5–20:1–2:1 indicating a strongly constrictional strain
362
isoclinal, intrafolial folds (F1 ) that refold S0 (Fig. 5b),
it is henceforth referred to as S0 /S1 . The most prominent fabric element in these mylonites is a penetrative
stretching lineation (L1 ) that plunges consistently
at moderate- to steep angles (45–75◦ ) to the ESE
Fig. 5. (a) Mineral rodding (L > S), here defined by positively weathering quartz rods, in gneisses of the Stolzburg pluton, ca. 100 m south
of the granite–greenstone contact on the farm Vergelegen (along the section of Fig. 6); oblique view normal to the lineation. (b) Oblique
view (looking W) of an isoclinal F1 fold that transposes bedding (S0 ) into the mylonitic S1 foliation. Folding is developed in felsic schist,
ca. 50 m north of the granite–greenstone contact on the farm Vergelegen (along the section of Fig. 6). (c) Moderately E- plunging (parallel to
hammer shaft) stretching lineation L1 developed in felsic schist on the farm Vereglegen, ca. 200 m north of the granite–greenstone contact.
trending mylonitic foliation that is subparallel to the
granite–greenstone contact and the S1 gneissosity in
gneisses of the Stolzburg pluton to the immediate
south (Fig. 6). Since this fabric is a transposition fabric that contains abundant centimetre- to metre-scale
363
zones. Over a distance of 300–400 m from the
granite–greenstone contact, amphibolite-facies mylonitic rocks are interleaved with greenschist-facies
rocks that grade into the low-grade central parts of
the SSB (Fig. 6). E–W trending, subvertical and
up to several metres wide quartz veins are abundant and coincide with the transition from amphibolite to predominantly greenschist-facies rocks in
the field. The gradual transition zone consists of
distinct, anastomosing ductile–brittle shear zones
characterized by greenschist-facies parageneses
along which the amphibolite-facies mylonitic fabrics are partially or completely replaced. The initially narrow greenschist-facies shear zones widen
into broader schist belts, made up of highly foliated
chlorite–albite–quartz and talc–carbonate–tremolite
schists. Significantly, there is a marked change in fabric development and greenschist-facies mylonites are
rather developed as S > L tectonites (Fig. 6) while the
prolate fabrics of amphibolite-facies rocks closer to
the granite–greenstone margin are no longer observed
although the SE-plunging L1 lineation is still prominent. The greenschist-facies mylonites and schists
contain abundant secondary shear foliations (see the
following), indicating a predominant non-coaxial
component of deformation. Metre-scale slivers of felsic schist and agglomerates contained in the predomi-
Fig. 6. Schematic cross-section (see Fig. 2 for location of section), illustrating the fabric development and structural relationships of
prograde and retrograde fabrics across the southern granite–greenstone contact (see text for detailed discussion).
throughout the southern margin of the SSB (Fig. 3b),
parallel to the L1 lineation developed in gneisses
of the Stolzburg pluton. The lineation is defined by
stretched and/or aligned minerals, and the rodding
of mineral aggregates, agglomerate fragments, quartz
veins, or felsic dykes. F1 fold axes show straight
hinge lines parallel to the L1 lineation. Sheath folds
that might be expected in these mylonite zones are
not observed. Although L–S tectonites are common
in the mylonitic contact zone, strain measurements
in, e.g. felsic schist units show axial ratios of agglomerate fragments and clasts of up to 30:1–2:1
indicating a strongly constrictional-type strain during fabric development. Shear sense indicators are
rare which, together with the predominantly prolate
fabrics in mylonites, suggests a rather bulk coaxial
strain during deformation. Primary intrusive contacts
between the Stolzburg pluton and the greenstone
sequence are indicated by trondhjemitic apophyses
within the greenstone sequence that have invariably
been transposed into the S0 /S1 fabric.
The strain intensity decreases away from the
granite–greenstone contact and primary bedding is
generally preserved within 200–300 m from the contact. S0 and S1 are parallel (Fig. 6), but fold transposition is only observed in metre-wide, E–W trending,
subvertical to steep southernly dipping high-strain
364
Greenstones along the northern margin of the SSB
are bounded by trondhjemitic gneisses of the 3236 ±
1 Ma Nelshoogte pluton (de Ronde and Kamo, 2000).
The supracrustal sequence is dominated by amphibolites and ultramafic rocks that include massive serpentinites and talc–carbonate schists (Fig. 2). Felsic schist
units that are characteristic for the southern margin
of the SSB are absent. The early transposition of fabrics (S0 /S1 ) is only evidenced by rootless, intrafolial
folds in banded amphibolites. Massive serpentinites,
in contrast, show little evidence of macroscopic tectonic fabrics. The L1 lineation is well developed in amphibolites being defined by the preferred orientation
of hornblende or stretched ocelli. In more siliceous,
possibly metasedimentary units, L1 is expressed by a
strong rodding of mineral aggregates. L1 plunges at
shallow- to steep angles to the W and WSW (Fig. 3c),
i.e. almost perpendicular to the easterly plunging L1
lineation in supracrustals along the southern margin
of the SSB. The contact between these two lineation
domains is sharp and occurs over a distance of only
50–100 m in the central part of the SSB (Fig. 3).
Contacts between the SSB and the Nelshoogte pluton to the north are sharp. Within several hundred metres to the granite–greenstone contact, the Nelshoogte
pluton typically shows a pervasive gneissose banding and foliation, but rodded textures as they are developed in the Stolzburg pluton to the south are not
preserved. The marginal gneissosity and banding of
the Nelshoogte pluton trends E–W and dips steeply
to the south, parallel to the mylonitic S0 /S1 fabric in
greenstones of the the SSB (Fig. 3d). However, mineral stretching lineations in the gneisses consistently
plunge at moderate angles to the E, i.e. approximately
3.4. The northern margin of the SSB
crenulation cleavage (Fig. 7a and b) or metre-scale
extensional duplexes (Fig. 7c). Shear sense indicators consistently point to a Stolzburg pluton-up
supracrustal-down sense of movement with a dextral strike-slip component. This oblique sense of
movement is consistent with the ESE plunge of the
penetrative L1 stretching lineation and the orientation
of extensional crenulation cleavages normal to the
L1 stretching lineation suggests that the lineation and
shear bands formed at the same time and within an
overall extensional shear zone.
nantly mafic- to ultramafic sequence testify to the intense structural disruption in these high-strain zones.
Amphibolite-facies mineral assemblages and fabrics
are preserved in boudins, metre-scale low-strain pods
and even laterally relatively continuous units as much
as 600 m away from the southern granite–greenstone
contact.
Amphibolites and minor greenschists along the eastern extent of the granite–greenstone contact lack mylonitic fabrics that are pervasively developed along the
western extent of the SSB. The lack of high-strain
fabrics in this area coincides with the undeformed nature of the Stolzburg pluton and the clearly intrusive
contacts of the trondhjemites into the supracrustal sequence. Mylonitic L–S tectonites are only developed
over a broad zone in the south-central part of the SSB
indicating that high-strain deformation is no longer
confined to the granite–greenstone contact, but that it
cuts up into the supracrustal sequence along the eastern extent of the SSB where the SSB merges with the
main BGB. The upper contact of this high-strain zone
coincides with a several hundred metres wide unit of
intensely foliated and lineated felsic schist (Fig. 2)
and agglomerates, typical of the Theespruit Formation of the Onverwacht Group, that is in contact with
strongly foliated chlorite schist. Very low-grade metamorphic rocks to the north of this zone only show a
widely spaced fracture and/or pencil cleavage with no
evidence of a tectonic lineation. Rocks to the south
of the high-strain zone are massive, though foliated
and lineated amphibolites, with minor occurrences of
talc–chlorite schist.
3.3. Kinematics of mylonites along the southern
granite–greenstone margin
Asymmetrical structures that could be used as
shear sense indicators are rare, though present, in
strongly lineated rocks, i.e. in rodded gneisses of the
Stolzburg pluton and amphibolite-facies mylonites
of the SSB. However, S–C fabric relationships are
common in greenschist-facies mylonites together
with occasional ␴- and rarer ␦-type porphyroclasts
and rotated fragments in, e.g. felsic schists and agglomerates. The S–C fabrics are developed on a
microscopic to metre-scale and micaceous units such
as quartz–sericite or chlorite schists locally contain a macroscopically well-developed extensional
365
of the SSB zone, shear sense indicators in mafic
and ultramafic rocks are rare. Moreover, the limited
macroscopic S–C fabric relationships point to a more
complex kinematic history, indicating predominantly
(1) a Nelshoogte pluton-up, supracrustal-down sense
of movement with a dextral strike-slip component, (2)
a less common dextral strike-slip movement only, and
(3) a rare supracrustal-up, Nelshoogte pluton-down
Fig. 7. S–C fabrics in greenstones along the southern margin of the SSB, consistently pointing to a combined N-down, S-up and
dextral strike-slip sense of movement. (a) S–C fabric in felsic quartz-sericite schist, ca. 50 m north of the granite–greenstone contact;
oblique view to the W. (b) Extensional shear bands in partly retrogressed banded ampibolite, ca. 250 m north of the granite–greenstone
contact, cross-sectional view, looking W. (c) Metre-scale extensional shear bands in retrograde chlorite schist, ca. 400 m north of the
granite–greenstone contact close to the Komati River north of the Vergelegen farmhouse; oblique view, looking W.
perpendicular to the westerly plunging L1 lineation in
the greenstones (Fig. 3c and d).
3.5. Kinematics of mylonites along the northern
margin of the SSB
Although the degree of mylonitization is not appreciably different to that of the southern margin
366
Peak pressure–temperature conditions can be constrained via the average P–T calculation method of
the programme THERMOCALC using the internally
consistent dataset of Holland and Powell (1998). Pressure estimates using this approach are consistent with
the results calculated for individual samples using
the conventional Grt-Hbl-Plg-Qtz barometer of Kohn
and Spear (1991) and the Grt-Hbl geothermometer
of Graham and Powell (1984). The results, calculated for a range of H2 O:CO2 mixtures in the fluid
phase, for An1, a garnet + plagioclase +hornblende
+ biotite + quartz rock and PB3, a garnet + plagioclase
+ hornblende + biotite + quartz + epidote + calcite
rock, are summarized in Table 2. Assuming an almost
pure water fluid composition, average P–T calculations for An1 and PB3 reveal peak-metamorphic P–T
conditions of 491 ± 40 ◦ C and 5.5 ± 0.9 kbar, and
492 ± 40 ◦ C and 6.3 ± 1.5 kbar respectively. As only
two out of the five independent reactions used to constrain peak conditions in An1 involve the presence
of a free fluid phase (e.g. they are devolatilisation
reactions), the P–T estimates are relatively insensitive
to changes of the water activities and the differences
between the estimated temperatures and pressures for
4.2. Conditions of metamorphism
with the development of later shear fabrics (S1 in
retrograde mylonites) that clearly postdate the peakmetamorphic porphyroblasts (Fig. 8A). These features
suggest a primary bulk-compositional control on the
distribution of the garnet-bearing peak-metamorphic
assemblages, and that these assemblages are probably
metamorphic grade equivalents of the predominant
peak assemblage in the amphibolites. Petrographic
and bulk-rock chemical data indicate that both the
presence of carbonate minerals in the amphibolites and
relatively high Fe/Fe + Mg ratios in the predominantly
magnesian amphibolites favour the development of
garnet. Retrogression is marked by the development
of assemblages consisting of actinolite + epidote +
chlorite + quartz in the metamafic rocks and muscovite + chlorite + quartz in the metapelitic layer. The
mineral chemistry of the garnet-bearing samples has
been investigated in detail (Dziggel, 2002) and representative mineral analyses of the peak assemblages
used to constrain peak-metamorphic conditions are
listed in Table 1.
sense of shear with a sinistral component. No consistent overprinting relationships could be identified that
would allow to establish a sequence of events.
3.6. F2 folding and steepening of lithologies
Throughout the SSB, earlier fabrics (i.e. S0 , S0 /S1 ,
L1 ) have been refolded by open to tight, upright folds
(F2 ). Folds range from centimetres to several tens of
metres in wavelength and amplitude and plunge at
moderate angles (25–45◦ ) to the E, typically somewhat
shallower than the earlier F1 transposition folds and
the L1 stretching lineation (Fig. 4c). The F2 folds show
invariably Z-shaped asymmetries along the southern
margin, S-shaped asymmetries along the northern margin and are mainly symmetrical in the central parts
of the SSB (Fig. 4d and e). Although lithologies can
not be matched across the belt, the orientation and
asymmetry of small-scale F2 folds indicate that the
SSB represents, at least in its westernmost parts, an
upright, isoclinal, shallow easterly-plunging synform.
It is this F2 folding that is responsible for the steepto subvertical dips of lithologies and tectonic fabrics
throughout the SSB (Fig. 4b). A subvertical, easterly
trending crenulation cleavage (S2 ) is well developed
in openly folded chlorite schists in the central parts of
the SSB and has an axial planar orientation to the F2
folds (Fig. 4b).
4. Metamorphism
4.1. Metamorphic assemblages and the relationship
of metamorphism to deformation
Throughout the study area, the predominant peakmetamorphic assemblage in the foliated amphibolites
of the Theespruit Formation is hornblende + plagioclase + sphene + quartz. Other locally developed assemblages useful in constraining peak-metamorphic
conditions are garnet + hornblende + plagioclase +
sphene + quartz, and garnet + plagioclase + hornblende + calcite + biotite + epidote + quartz in metamafic rocks; and garnet + biotite + muscovite + quartz
in a single metapelitic layer. All the garnet-bearing
assemblages are confined to specific narrow layers
developed parallel to the compositional banding of
the rocks (S0 ). In all cases retrogression is associated
367
Fig. 8. Typical mineral textures developed in (A) the garnet and calcite bearing amphibolite (PB3) and (B) the metapelitic layer. In (A)
the S1 floliation defined by hornblende (Hbl) wraps around a poikiloblastic garnet (Grt) porphyroblast developed in a calcite-bearing (Cc)
band within the amphibolite. Quartz and calcite inclusions within the garnet porphyroblast are aligned to an older foliation. (B) Illustration
of the partial replacement of a peak-metamorphic biotite (Bt) by muscovite (Mus) and chlorite (Chl) along the S1 foliation. In both cases
the scale bar represents 1 mm.
Hbl
41.63
0.48
17.77
15.61
0.39
7.82
11.41
1.40
0.46
0.03
0.24
0.100
97.35
6.233
1.767
1.366
0.054
1.745
1.834
0.12
0.049
1.83
0.406
0.088
15.49
0.47
SiO2
Al2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al
Fe2+
Mn
Mg
Ca
Na
K
Total
XAn
XAb
Pl
55.3
28.6
0.1
0.00
0.00
10.1
6.4
0.1
100.57
9.922
6.043
0.015
0.000
0.000
1.942
2.227
0.023
20.172
0.47
0.53
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
ZnO
ZrO2
Total
Si
AlIV
AlVI
Ti
Fe2+
Cr
Mn
Mg
Ca
Na
K
Total
Xfe
Bt
31.05
0.32
21.97
0.11
22.69
0.85
11.68
0.04
0.15
8.70
0.50
0.03
97.07
5.089
2.911
1.329
0.039
3.11
0.014
0.118
2.854
0.007
0.048
1.676
15.874
0.521462
PB3
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
Total
TSi
TAl
AlVI
Fe3+
Ti
Cr
Fe2+
Mg
Mn
Ca
Na
Total
XGrs
XAlm
XPyp
XSpss
Grt
35.71
0.10
20.79
0.00
3.36
25.97
5.86
0.84
7.33
0.00
99.96
5.794
0.206
3.770
0.410
0.012
0.000
3.524
0.204
0.806
1.276
0.000
8.000
0.22
0.66
0.04
0.14
SiO2
TiO2
Al2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Cl2 O
ZnO
ZrO2
Total
TSi
TAl
CAl
Fe3+
CTi
CMg
CFe2+
BMn
BCa
ANa
AK
Total
XMg
Hbl
40.27
0.34
19.98
2.66
17.40
0.50
4.46
11.58
0.76
0.80
na
0.00
0.00
98.76
5.988
2.012
1.491
0.298
0.038
0.988
2.165
0.063
1.845
0.219
0.152
15.259
0.31
SiO2
Al2 O3
FeO
CaO
Na2 O
K2 O
Total
Si
Al
Fe2+
Ca
Na
K
Total
XAn
XAb
Pl
55.84
28.62
0.00
10.57
5.53
0.01
100.57
9.984
6.032
0.00
2.024
1.916
0.004
19.96
0.51
0.49
SiO2
TiO2
Al2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
AlIV
AlVI
Ti
Fe3+
Fe2+
Mn
Mg
Ca
Na
K
Total
XFe
Bt
34.25
1.74
18.58
3.39
21.70
0.20
7.92
0.01
0.14
8.82
96.76
5.244
2.756
0.598
0.200
0.390
2.778
0.026
1.808
0.002
0.042
1.724
15.568
0.39
SiO2
TiO2
Al2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al
Ti
Fe3+
Fe2+
Mn
Mg
Ca
Na
K
Total
Ep
38.26
0.21
28.69
4.28
3.43
0.38
0.00
25.13
0.20
0.23
100.81
2.937
2.596
0.012
0.247
0.220
0.025
0.000
2.067
0.030
0.023
8.156
368
Grt
37.56
0.03
21.22
0.00
25.06
6.76
1.66
7.62
0.12
100.0
6.01
0.00
3.998
0.021
0.004
0
3.332
0.396
0.916
1.306
0.037
16.02
0.219
0.56
0.067
0.153
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Cl2 O
ZnO
ZrO2
Total
TSi
TAl
CAl
CTi
CMg
CFe2+
BFe2+
BMn
BCa
ANa
AK
Total
XMg
Table 1
Representative analyses for the minerals defining the peak-metamorphic assemblages from An1 and PB3
An1
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
Total
Tsi
Tal
AlVI
Fe3+
Ti
Cr
Fe2+
Mg
Mn
Ca
Na
Total
XGrs
XAlm
XPyp
XSpss
The structural formulae were calculated on the basis of 24 oxygen for garnet, 23 oxygen for amphibole, 32 oxygen for plagioclase, 22 oxygen for biotite, and 13 oxygen
for epidote. Fe3 = has been recalculated by charge balance. The carbonate forming part of the peak assemblage in PB3 is a mixture of 93.6% calcite, 1.8% magnesite and
3.4% siderite.
Sample
XH2 O
491
483
473
458
T
40
37
63
40
39
37
35
S.D.
(T)
6.3
8.0
14.1
5.5
5.4
5.3
5.1
P
1.5
1.3
2.3
0.9
0.9
0.9
0.9
S.D.
(P)
0.922
0.904
0.921
0.605
0.601
0.595
0.589
Correlation
value
2.88
2.46
3.14
0.85
0.97
0.91
0.98
Fit
369
370
phibolite from which the zircons were extracted contained a metamorphic mineralogy and texture indistinguishable from the bulk of the Theespruit Formation
amphibolites from the study area. Thus, irrespective of
the origin it would appear to date amphibolite-facies
metamorphism at 3219 Ma or younger.
5.1. The structural significance of granite–greenstone
contacts of the SSB
schists at higher stratigraphic levels. Notably, the
L1 lineation trajectories describe an unidirectional
NE–SW trend throughout the supracrustals of the
SSB and in gneisses of the Stolzburg pluton in this
pre-F2 orientation.
The strain intensity varies throughout the SSB.
Most significantly, however, there is a marked
change in the strain regime from footwall gneisses
and amphibolite-facies mylonites at and below the
granite–greenstone contact to greenschist-facies shear
zones higher up in the succession. Prolate fabrics
dominate in rodded gneisses of the Stolzburg pluton and amphibolite-facies mylonites. The pervasive
L-fabrics in these rocks, together with the lack of
asymmetric foliations and shear sense indicators, indicate a bulk coaxial strain during fabric development.
The high-grade metamorphic constrictional fabrics
record, in a subhorizontal orientation, a vertical shortening and horizontal NE–SW stretching of the deeper
structural levels at and below the granite–greenstone
contact. The prominent L1 linear fabric developed
throughout the Stolzburg pluton, albeit at lower strain
intensities compared to the gneisses close to the
Fig. 10. U–Pb Concordia diagram for the zircon bearing amphibolite. Errors are at one sigma. The inset SEM backscattered electron image
illustrates the nature of the zoning in the zircon grains extracted from this sample as well as the zone that was analysed.
4.3. Age of peak metamorphism
The stuctural significance of the highly deformed
granite–greenstone contacts along the SSB can only
be discussed once the sequence has been restored to its
original orientation, and assuming an initially subhorizontal attitude, the upright lithologies and fabrics have
to be rotated about the shallow easterly plunging F2
fold axes. Upon unfolding, gneisses of the Stolzburg
pluton form the footwall of the supracrustal sequence
and are successively overlain by amphibolite-facies
and lower grade greenschist-facies mylonites and
5. Discussion
An attempt was made to constrain the age of peak
metamorphism by dating zircons from the metamafic
rocks. A single metamafic layer yielded four zircon
grains. The grains are very small (ca. 50 ␮m) and
record a distinct concentric compositional zoning pattern. One grain was analysed on SHRIMP II at the
Research School for Earth Sciences (RSES) at the
Australian National University, Canberra, and the result is shown in Fig. 10. The analytical procedure and
details of the elemental and isotopic data are presented
in Dziggel (2002). The zircon is relatively rich in Th
(102 ppm) and has a high Th/U ratio of 0.98. The analysis is near concordant, and reveals a 207 Pb/206 Pb date
of 3219±9 Ma. The high Th/U ratio, together with the
concentric zoning pattern, may indicate a magmatic
rather than metamorphic origin. Despite this the am-
dehydration reactions and are consequently sensitive
to fluid compositions estimations. Water-dominated
fluids produce P–T conditions consistent with those
derived for An1, whilst CO2 -rich fluid produce significantly higher P–T estimates (Table 2 and Fig. 9).
In the area investigated, the garnet- and carbonatebearing layers occur as narrow bands within amphibolites. This, as well as the close spatial association
of An1 and PB3, suggests that water-rich fluids
dominated in both rocks. In general, the relatively
low-temperature estimates are in good agreement
with the presence of epidote inclusions in garnet and
hornblende in An1 and the existence of epidote in the
peak-metamorphic assemblage of PB3.
The pressure–temperature conditions of equilibration for greenschist-facies mineral assemblages in
metamafic rocks are difficult to constrain as no reliable
geobarometers or cation exchange geothermometers
are available for this system. Retrograde assemblages
in the metamafic rocks of the study area are typified
by lower greenschist-facies mineral assemblages and
in the metapelitic layer, peak-metamorphic biotite is
clearly replaced by a retrograde muscovite + chlorite
assemblage (Fig. 8B). Within the pressure range
2–4 kbar, muscovite and chlorite breakdown to form
biotite at maximum temperature of approximately
400 ◦ C (Ferry, 1984) indicating that deformation and
retrogression in the study area persisted to below
400 ◦ C.
An1
1.0
0.8
0.5
0.3
492
557
716
Table 2
Thermocalc average P–T results for the peak assemblages in An1
and PB3
PB3
0.9
0.5
0.1
the respective water activities are in fact smaller than
the absolute errors calculated pressure and temperature (Table 2 and Fig. 9). Peak conditions in sample
PB3 are constrained by a range of decarbonation and
Fig. 9. Average P–T conditions with errors calculated via
THERMOCALC for An1 and PB3. The likely garnet producing reaction in both rocks, i.e. 21An + 6Fact = 11 Gr + 10Alm +
27Q + 6H2 O in An1 and 24An + 3Fact + 5Parg + 4Mag = 5Ab +
8Gr + 5Alm + 8Ts + 4CO2 in PB3 are also indicated. The shaded
area for the latter reaction represents the shift in the P–T conditions of the reaction resulting from change in fluid compositions
indicated for sample PB3.
371
deformed amphibolite-facies rocks by retrograde mylonites illustrates the fact that greenschist-facies mylonites progressively cut down into Lower Plate rocks
during the exhumation of the basement rocks. The
widespread preservation of amphibolite-facies relics
in greenschist-facies schists testifies to the originally
much wider extent of Lower Plate rocks in the the
SSB and suggests that significant parts of the greenstone sequence along the granite–greenstone contacts
have been lost during extensional shearing.
The greenschist-facies mylonites and schist zones
in the SSB, thus, represent the actual extensional detachment that separates Lower Plate from Upper Plate
rocks in the metamorphic core complex (Fig. 11).
Lower Plate rocks include much of the southern gneiss
terrain with the high-pressure rocks in the Stolzburg
pluton described by Dziggel et al. (2002) forming
the deepest parts that have so far been identified. The
high-grade metamorphic and partially retrogressed
rocks at the base of the BGB that largely correspond
to rocks of the Theespruit Formation are also part
of the Lower Plate (Fig. 11). Notably, the relatively
high-P/low-T metamorphic conditions recorded in
these rocks are of the same type, albeit at lower
grades, as those documented by Dziggel et al. (2002)
in the Stolzburg pluton. Upper Plate rocks include the
low- and very low-grade metamorphic rocks above
the detachment. These rocks are only exposed in the
easternmost parts of the SSB in the main part of
the BGB.
The extensional detachment is not confined to
the granite–greenstone margin, but cuts up into
the supracrustal succession along its eastern extent
(Fig. 2). The high-grade metamorphic and variably
retrogressed rocks below the detachment, e.g. in the
Tjakastad schist belt are characterized by predominantly constrictional fabrics (Kisters and Anhaeusser,
1995) that are typical for the Lower Plate rocks.
The eastern strike continuation of the Lower Plate
rocks corresponds to the melange-like units of the
Theespruit Formation. The Theespruit Formation is
bounded in the north against the low-grade rocks of
the Komati Formation by the Komati Fault that has
traditionally been regarded as a thrust fault (e.g. de
Wit et al., 1987; Armstrong et al., 1990) although no
detailed structural and kinematic data have been been
presented for the fault. This work suggests that the
Komati Fault represents the eastern strike extent of
granite–greenstone contacts, implies a penetrative
ductile flow of the trondhjemitic basement gneisses
during NE–SW stretching. Greenschist-facies mylonites and schist zones, in contrast, are typically S >
L tectonites. They are developed as distinct and, initially, relatively narrow shear zones that progressively
replace the amphibolite-facies mylonites towards the
central, structurally higher parts of the SSB. Asymmetric shear foliations are common in these mylonites and
fabric development indicates that the retrograde mylonites have formed during predominantly non-coaxial
deformation. Kinematic indicators consistently point
to a top-to-the-NE sense of movement of the greenstone sequence relative to the underlying Stolzburg
pluton, and parallel to the horizontal NE–SW stretching of amphibolite-facies supracrustals and gneisses in
the footwall. The widespread extensional structures in
the mylonites suggest that shearing occurred along an
extensional shear zone. This agrees with the overprint
of amphibolite-facies rocks by greenschist-facies mylonites indicating that shearing has occurred during
cooling and probably uplift of the rocks.
This spatial relation between the constrictional-type,
coaxial strain regime in high-grade rocks at and below
the granite–greenstone contacts overprinted by rather
plane-strain/flattening-type non-coaxial deformation
in greenschist-facies mylonites at higher structural
levels is characteristic for the fabric development in
extensional detachments associated with the extensional collapse of orogens (e.g. Dewey, 1988; Davis,
1988; Reynolds and Lister, 1990; Malavielle and
Taboada, 1991; Hill, 1994; Andersen et al., 1994;
Krabbendam and Dewey, 1998). In this scenario,
the high-grade metamorphic prolate fabrics at the
base of the BGB are collapse-related fabrics that
formed in response to crustal thickening, i.e. vertical
shortening and horizontal stretching that is accommodated by penetrative, coaxial ductile flow at these
mid- and lower crustal levels (e.g. Dewey, 1988). In
upper-crustal rocks, the horizontal stretching and thus
thinning of the crust is accommodated along more
discrete extensional shear zones and/or normal faults.
These shears are characterized by predominantly rotational, simple shear deformation. Horizontal stretching and coaxial flow of mid-crustal levels is, in all
likelihood, contemporaneous with the non-coaxial
deformation in retrograde shear zones at upper crustal
levels. The progressive replacement of the coaxially
372
to the east of the SSB and that marks the cessation of
regional tectonism in the area (de Ronde and de Wit,
1994; Kamo and Davis, 1994; Lowe et al., 1999).
Given this narrow age bracket for the exhumation of
the TTG terrain, extensional detachment faulting must
have occurred during or shortly after the main D2
collisional event at ca. 3230–3225 Ma. The NE–SW
directed stretching of the mid- and lower crustal
rocks and top-to-the-NE sense of movement along
the extensional detachment are perpendicular to the
NW-directed thrusting inferred for the D2 compressional event (e.g. Lamb, 1984; de Wit et al., 1987; de
Ronde and de Wit, 1994; Kamo and Davis, 1994; de
Ronde and Kamo, 2000) which indicates the lateral
extrusion of mid- and lower crustal levels.
On a regional scale, the ca. 3225 Ma timing of
extensional collapse coincides with the onset of sedimentation of the up to 3500 m thick sequence of
coarse-clastic, terrigeneous Moodies Group that is
commonly believed to be derived from the erosion
of the uplifted surrounding TTG terrain (e.g. Jackson
et al., 1987; Lowe, 1999). Recently, Heubeck and
Lowe (1994) and Lowe (1999) have proposed an extensional setting of Moodies Group deposition in a
number of restricted, probably fault-bounded basins,
rather than sedimentation in a foreland basin (Jackson
Fig. 11. Schematic outline of the Stolzburg schist belt illustrating the different parts of the extensional detachment exposed in and adjacent
to the schist belt. Lower Plate rocks include the high-grade metamorphic marginal zones of the greenstone belt as well as the Stolzburg
pluton to the south. The mylonitic front of the detachment is characterized by retrograde greenschist-facies shear zones, that cut into Lower
Plate rocks during exhumation. Upper Plate rocks include the very low-grade rocks of the central parts of the BGB. The Nelshoogte pluton
to the north is an early- to synkinematic pluton; the role of the Badplaas pluton in the west is not known at this stage. The 3.1 Ga Kees
Zyn Doorns syenite is post-tectonic.
the extensional detachment identified in the SSB, an
aspect that awaits further detailed work.
5.2. Timing of extensional collapse
There are several lines of evidence that point to an
orogenic collapse of the BGB during or slightly after
the main D2 collisional event at ca. 3230–3225 Ma.
The D2 deformation is the main collisional event in
the BGB (de Ronde and de Wit, 1994; Kamo and
Davis, 1994) and considering that peak-metamorphic
conditions and associated burial of the southern
granite-gneiss terrain were only attained at ca.
3230 Ma (Dziggel et al., 2002; Stevens et al., 2002),
the exhumation of the southern TTG terrain and its
juxtaposition with the BGB along the extensional
detachment must have occurred after 3230 Ma. A
likely age bracket for the high-grade metamorphism
in greenstones of the Lower Plate is provided by the
3219 ± 9 Ma zircon age from a deformed amphibolite
along the southern margin of the SSB (Fig. 10). On a
regional scale, an upper age bracket for exhumation
tectonics is given by the intrusion of the post-tectonic,
3216 + 2/−1 Ma Dalmein pluton (Kamo and Davis,
1994) (Fig. 1) that sharply truncates all earlier structures along the southern granite–greenstone margin
373
BGB (e.g. de Ronde and Kamo, 2000). In contrast, our
observations rather suggest a late-stage steepening of
pre-existing fabrics during solid-state diapirism of the
surrounding granite gneisses. Notably, regional-scale
cleavage triple points are developed in proximity to
the bordering granitoids in, e.g. the Nelshoogte schist
belt to the immediate north of the SSB (Anhaeusser,
2001) that, in the absence of any cross-folding, point
to the steepening of fabrics during a solid-state diapiric component of emplacement of the surrounding
granite-gneiss terrane. The late-stage diapiric component may have initiated due to isostatic instabilities
set up during the crustal thinning associated with the
extensional collapse of the BGB, a well-documented
feature of many Phanerozoic core complexes that
leads to the formation of basement culminations (e.g.
Lister and Davis, 1989) (Fig. 12). Considering the exhumation of hot basement rocks and the voluminous
synkinematic 3.225 Ma TTG plutonism, particularly
along the northern margin of the BGB, the rheological
contrast between hot and buoyant basement and intrusive TTG’s and dense, cool, overlying supracrustals
of mainly mafic and ultramafic composition may have
triggered the transition to the actual solid-state diaprism. This multistage evolution from early crustal
thinning and associated core-complex formation to
subsequent solid-state diaprism accentuated by density inversions (Fig. 12) has been proposed by, e.g.
Marshak et al. (1992, 1997) and Marshak (1999) to
account for the typical dome-and-keel pattern of many
Archaean and Paleoproterozoic granite–greenstone
provinces. Along the southern granite–greenstone
margin of the SSB mylonitic fabrics and condensed
metamorphic gradients that were previously considered to be related to the diapiric emplacement of
adjoining plutons (e.g. Viljoen and Viljoen, 1969;
Anhaeusser, 1984, 2001) have largely formed prior to
doming. Fabrics related to the late-stage steepening
of lithologies are, however, prominently developed
in the central parts of the SSB, where the later,
penetrative S2 axial planar foliation formed at high
angles to the earlier, extension-related fabrics and
bedding. It seems likely that these steepening-related
fabrics are more widespread in the SSB. However, a
clear distinction between the earlier S1 and later S2
fabrics remains ambiguous due to the coplanar orientation of the two in the subvertical, highly strained
marginal zones of the SSB. The presence of shear
et al., 1987). These observations are all consistent
with a deposition of the Moodies Group in, e.g. extensional half-graben or graben structures that developed in the collapsing hanging wall of the evolving
extensional detachment.
5.3. The late-stage steepening of fabrics
The present-day, subvertical attitude of virtually
all lithologies and fabrics along the SSB and, in fact,
the bulk of the BGB is clearly at variance with the
commonly shallow dips of low-angle detachments
found in, e.g. Phanerozoic metamorphic core complexes. The steepening of lithologies and fabrics in
the BGB is commonly thought to have occurred during the progressive NW–SE directed D2 shortening
of the greenstone sequence (e.g. de Wit et al., 1987;
Lowe, 1999). Similarly, the often arcuate geometry of
fabrics in proximity to the dome-like TTG plutons is
interpreted to be a consequence of the ‘moulding’ of
regional-scale structures around the rigid plutons during progressive D2 shortening (e.g. Tomkinson and
King, 1991; Lowe, 1999; Lowe et al., 1999). Although
the progressive steepening of initially low-angle structures during bulk shortening is a well-documented
feature of many fold- and-thrust belts, it can hardly
account for the subvertical granite–greenstone contacts developed along the entire margin of the BGB
without invoking a vertical component of movement
of the TTG’s relative to the BGB.
The orientation and symmetry of F2 folds in the
SSB, the associated upright, E-trending axial planar
S2 cleavage and the absence of transsecting cleavages
together with the coplanar and colinear orientation of
mylonitic fabrics in supracrustals of the SSB and the
Stolzburg pluton are difficult to reconcile with a mere
refolding of early NE–SW trending D2 structures
around rigid TTG plutons. Moreover, the Stolzburg
pluton is bounded by the N–S trending Tjakastad
schist belt in the east (Fig. 1) that is characterized by
pervasive N–S trending lithologies and prolate fabrics.
The Stolzburg and Tjakastad schist belts trend at right
angles to each other and there is no evidence that the
orthogonal trend of the belts is caused by the refolding
of regional fabrics. It is also clear that the pervasive
prolate fabrics in both the Stolzburg and Tjakastad schist belts bear no resemblance to the mainly
plane-strain to flattening-type fabrics within the main
374
flects the steepening and, thus, backfolding of the
top-to-the-NE extensional detachment in the north.
In this study, we have documented the fabric development along granite–greenstone contacts in and
around the SSB that is mainly related to the ca. 3.2 Ga
exhumation of rocks. Evidence of fabrics associated
with the 3230 Ma burial or even the 3445 Ma D1
fabrics related to diaprism may be indicated by the
complexity of shear sense indicators along the northern margin of the SSB compared to the uniform
kinematics recorded in mylonites along the southern margin. The presence of both greenstone-up as
well as greenstone-down kinematic indicators along
the northern granite–greenstone contacts possibly re-
䉳
375
The structural and metamorphic evolution of the
southern granite–greenstone margin of the SSB indicates that the highly tectonized and lithologically
heterogeneous granite–greenstone contacts form part
of an extensional detachment zone. Along this detachment, lower crustal TTG’s of the southern gneiss terrain have been juxtaposed against the very low-grade
metamorphic supracrustal rocks of the central parts
of the BGB. The progressive overprint of high-grade
coaxial fabrics of rocks at deeper structural levels by
retrograde non-coaxial fabrics in the detachment zone,
the top-to-the-NE kinematic indicators in retrograde
detachment mylonites, together with the sharp metamorphic break along the granite–greenstone margins
indicate that detachment faulting and exhumation of
the high-grade TTG terrain developed in response to
the extensional orogenic collapse of the BGB. The
propagation of the detachment shear zones into deeper
crustal rocks during exhumation results in the juxtaposition of two different structural and, thus, meta-
5.4. Conclusions
by relatively rare oblique sinistral supracrustal-up,
Nelshoogte pluton-down shear sense indicators along
the northern margin of the SSB. Most shear sense
indicators rather point to oblique off-the-dome kinematics which are possibly related to the later diapiric
steepening of fabrics. At this stage, our data set is
insufficient to provide a conclusive answer as to the
emplacement mechanisms and kinematics of synkinematic plutons along the northern margin of the BGB
which is also beyond the scope of this paper.
event in the southern part of the BGB appear to have
been largely destroyed and overprinted during exhumation tectonics. However, these older fabrics are
most likely preserved in the eastern parts of the SSB
where the extensional detachment cuts up section and
into the greenstone belt so that amphibolite-facies
mafic and ultramafic schists are preserved structurally below the detachment. The supracrustals are
in contact with virtually undeformed trondhjemites
of the intrusive Stolzburg pluton and the truncation
of amphibolite-facies fabrics by the Stolzburg pluton suggests that these fabrics pre-date the 3445 Ma
intrusion of the trondhjemites.
Our model of deep-crustal exhumation in response
to an extensional collapse of the BGB at ca. 3220 Ma
implies that the response of the southern, high-grade
metamorphic TTG terrain to exhumation and uplift is likely to be different to that of the 3236 Ma
Nelshoogte pluton along the northern margin of the
SSB. The Nelshoogte pluton represents a largely
early- to syn-D2 pluton and although it contains a
pervasive gneissosity along its margins, it has not presented a lower crustal basement to the supracrustals
as the older Stolzburg pluton in the south. This
most likely explains the markedly different fabric
development in the Nelshoogte and Stolzburg plutons, namely the lack of constrictional-type fabrics
in gneisses of the Nelshoogte pluton, and the lineation pattern, that deviates from the unidirectional
L1 lineation pattern in the SSB and Stolzburg pluton.
Top-to-the-NE extensional shearing inferred for the
upper-plate greenstone sequence is, in the present-day
upright orientation of the greenstones, only manifest
Fig. 12. Cartoon of the evolution of core-complex formation and extensional detachment faulting in the southern Barberton granite–greenstone
terrain as envisaged in this study. (a) Crustal thickening during presumably NW-verging thrusting at ca. 3230–3225 Ma (e.g. de Ronde and
de Wit, 1994; Kamo and Davis, 1994). Both the supracrustal sequences as well as the basement are thickened and are intruded by earlyto syn contractional plutons such as the Nelshoogte pluton. (b) Schematic SW–NE crustal section, drawn parallel to the strike of the BGB
and perpendicular to the vergence of earlier thrusts shown in (a). The thickening of the crust results in vertical shortening and horizontal
stretching of lower crustal levels. The bulk coaxial flow and crustal extension at this deeper structural level is accommodated by non-coaxial
faulting and top-to-the-NE extension at upper crustal levels. (c) During tectonic denudation of upper crustal levels, high-grade metamorphic
basement rocks that have undergone peak-metamorphic conditions during the earlier contractional stage are buoyant relative to the cooler
and denser overlying supracrustal sequence that is made up of mainly mafic and ultramafic rocks. This initiates basement culminations,
typically found in core complex terrains. During this time and shortly following the contractional tectonics, syn-extensional plutons intrude
along the NE margin of the BGB (e.g. Kaap Valley tonalite at 3225 Ma). Sedimentation of intramontane molasse in half-graben structures
that develop above the collapsing hanging wall of the extensional detachment are represented by the onset of Moodies Group deposition
(stippled) at the same time at ca. 3225 Ma. (d) The basement culminations are accentuated to form solid-state diapirs. During the diapiric
stage, the extensional detachment and fabrics are steepened and the late-tectonic molasse deposits are folded. A reactivation of the initially
normal faults as thrust faults or vice versa is likely during this stage.
376
Anhaeusser, C.R., 1969. The Stratigraphy, Structure and Gold
Mineralization of the Jamestown and Sheba Hills Area of
the Barberton Mountain Land. Ph.D. Thesis. University of the
Witwatersrand, Johannesburg, South Africa, 322 pp.
Anhaeusser, C.R., 1984. Structural elements of Archaean
granite–greenstone terranes as exemplified by the Barberton
Mountain Land, South Africa. In: Kroener, A., Greiling, R.
(Eds.), Precambrian Tectonics Illustrated. E. Schweizerbart’sche
Verlagsbuchhandlung, Stuttgart, Germany, 419 pp.
Anhaeusser, C.R., 2001. The anatomy of an extrusive-intrusive
Archaean mafic-ultramafic sequence: the Nelshoogte Schist
Belt and Stolzburg Layered Ultramafic Complex, Barberton
Greenstone Belt, south Africa. S. Afr. J. Geol. 104, 167–204.
Anhaeusser, C.R., Robb, L.J., Viljoen, M.J., 1981. Provisional
Geological Map of the Barberton Greenstone Belt and
Surrounding Terrane, Eastern Transvaal and Swaziland.
Geological Society of South Africa, Johannesburg, Scale
1:250,000.
Armstrong, R.A., Compston, W., de Wit, M.J., Williams, I.S., 1990.
The stratigraphy of the 3.5–3.2 Ga Barberton Greenstone Belt
revisited: a single zircon ion microprobe study. Earth Planet.
Sci. Lett. 101, 90–106.
Cloete, M., 1991. An overview of metamorphism in the Barberton
greenstone belt. In: Ashwal, L.D. (Ed.), Two Cratons and
an Orogen—Excursion Guidebook and Review Articles for
a Field Workshop Through Selected Archaean Terranes of
Swazioland, South Africa and Zimbabwe. IGCP project
280. Department of Geology, University of Witwatersrand,
Johannesburg, pp. 85–98.
Collins, W.J., Van Kranendonk, M.J., Teyssier, C., 1998. Partial
convective overturn in the east Pilbara Craton, Western
Australia: driving mechanisms and tectonic implications. J.
Struct. Geol. 20, 1405–1429.
Davis, G.A., 1988. Rapid upward transport of mid-crustal mylonitic
gneisses in the footwall of a Miocene detachment fault, Whipple
Mountains, southeastern California. Geologische Rundschau 77,
191–209.
de Ronde, C.E.J., Kamo, S.L., 2000. An Archaean arc-arc
collisional event: a short-lived (ca. 3 Myr) episode, Weltevreden
area, Barberton greenstone belt, South Africa. J. Afr. Earth Sci.
30, 219–248.
de Ronde, C.E.J., de Wit, M.J., 1994. Tectonic history of the
Barberton greenstone belt, South Africa: 490 million years of
Archean crustal evolution. Tectonics 13, 983–1015.
Dewey, J.F., 1988. Extensional collapse of orogens. Tectonics 7,
1123–1139.
de Wit, M.J., 1983. Notes on a preliminary 1:25,000 geological
map of the southern part of the Barberton greenstone belt. In:
Anhaeusser, C.R. (Ed.), Contributions to the Geology of the
Barberton Mountain Land. Geological Society of South Africa
Special Publication No. 9, pp. 185–187.
de Wit, M.J., 1998. On Archaean granites, greenstones, cratons
and tectonics: does the evidence demand a verdict? Precambrian
Res. 91, 181–226.
de Wit, M.J., Fripp, R.E.P., Stanistreet, I.G., 1983. Tectonic and
stratigraphic implications of new field observations along the
southern part of the Barberton greenstone belt. In: Anhaeusser,
C.R. (Ed.), Contributions to the Geology of the Barberton
morphic levels. This accounts for the abrupt metamorphic breaks that characterize the granite–greenstone
margins of the BGB and that have been noted by previous studies (e.g. Anhaeusser, 1984; Cloete, 1991).
A further consequence of core-complex formation is
that the southern, highly tectonized and complexely
metamorphosed southern margin of the BGB, that
largely corresponds to the Theespruit Formation, is
entirely allochthonous with respect to the main parts
of the BGB and the result of a tectonic underplating.
The timing of extensional detachment faulting at
the base of the BGB coincides with the onset of
the coarse-clastic Moodies Group sedimentation at
ca. 3225 Ma. The deposition of the thick sequence
of quartz-rich, terrigeneous sediments in small,
fault-bounded basins in the central parts of the BGB
(e.g. Heubeck and Lowe, 1994) is tentatively related
to the extensional collapse of hanging wall-rocks
above the detachment.
We suggest that the steepening of fabrics to their
present-day vertical attitudes reflects a late-stage isostatic instability caused by the horizontal stretching
and vertical thinning of crust during which less dense
and hot basement gneisses rose, at a late stage of deformation, as solid-state diapirs through the overlying denser, mafic to ultramafic supracrustal sequence.
As such, core complex and dome-and-keel formation
represent part of a continuum, similar to the multistage evolution of Archaean and Paleoproterozoic
dome-and-keel provinces proposed by, e.g. Marshak
et al. (1992, 1997).
Acknowledgements
This material is based upon work supported by the
National Research Foundation under Grant number
NRF2050238. We would like to thank all farmers and
land owners in the Badplaas and Tjakastad areas for
their hospitality and granting access to their lands.
Helpful reviews by K.E. Karlstrom and C.R.L. Friend
are gratefully acknowledged.
References
Andersen, T.B., Osmundsen, P.T., Jolivet, L., 1994. Deep crustal
fabrics and a model for extensional collapse of the southwest
Norwegian Caledonides. J. Struct. Geol. 16, 1191–1203.
377
Caledonides. In: Holdsworth, R.E., Strachan, R.A., Dewey,
J.F. (Eds.), Continental Transpressional and Transtensional
Tectonics. Geological Society of London Special Publication
No. 135, pp. 159–181.
Kröner, A., Byerly, G.R., Lowe, D.R., 1991. Chronology of
early Archaean granite-greenstone evolution in the Barberton
mountain land, South Africa, based on precise dating by singlezircon evaporation. Earth Planet. Sci. Lett. 103, 41–54.
Kröner, A., Hegner, E., Wendt, J.I., Byerly, G.R., 1996. The
oldest part of the Barberton granitoid-greenstone terrain, South
Africa, evidence for crust formation between 3.5 and 3.7 Ga.
Precambrian Res. 78, 105–124.
Lamb, S.H., 1984. Geology of Part of the Archaean Barberton
Greenstone Belt, Swaziland. Unpublished Ph.D. Thesis.
Cambridge University, England, 250 pp.
Lister, G.S., Davis, G.A., 1989. The origin of metamorphic
core complexes and detachment faults formed during Tertiary
continental extension in the northern Colorado River region,
USA. J. Struct. Geol. 11, 65–94.
Lopez-Martinez, M., York, D., Hall, C.M., Hanes, J.A., 1984.
Oldest reliable 40 Ar/39 Ar ages for terrestrial rocks: Barberton
Mountain komatiites. Nature 307, 352–354.
Lowe, D.R., 1994. Accretionary history of the Arcean Barberton
Greenstone Belt (3.55–3.22 Ga), southern Africa. Geology 22,
1099–1102.
Lowe, D.R., 1999. Geologic evolution of the Barberton Greenstone
Belt and vicinity. In: Lowe, D.R., Byerly, G.R. (Eds.),
Geologic Evolution of the Barberton Greenstone Belt, South
Africa. Geological Society of America Special Paper No. 329,
pp. 287–312.
Lowe, D.R., Byerly, G.R., 1999. Stratigraphy of the west-central
part of the Berberton Greenstone Belt, South Africa. In: Lowe,
D.R., Byerly, G.R. (Eds.), Geologic Evolution of the Barberton
Greenstone Belt, South Africa. Geological Society of America
Special Paper No. 329, pp. 1–36.
Lowe, D.R., Byerly, G.R., Heubeck, C., 1999. Structural divisions
and development of the west-central part of the Barberton
Greenstone Belt. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic
Evolution of the Barberton Greenstone Belt, South Africa.
Geological Society of America Special Paper No. 329,
pp. 83–114.
Malavielle, J., Taboada, A., 1991. Kinematic model for postorogenic Basin and Range extension. Geology 19, 555–558.
Marshak, S., 1999. Deformation style way back when: thoughts
on the contrasts between Archaean/Paleoproterozoic and
contemporary orogens. J. Struct. Geol. 21, 1175–1182.
Marshak, S., Alkmim, F., Jordt-Evangelista, H., 1992. Proterozoic
crustal extension and the generation of dome-and-keel structures
in an Archaean granite-greenstone terrane. Nature 357, 491–
493.
Marshak, S., Tinkham, D., Alkmim, F., Brueckner, H.,
Bornhorst, T., 1997. Dome-and-keel provinces formed during
Paleoproterozoic orogenic collapse—core complexes, diapirs,
or neither? Examples from the Quadrilatero Ferrifero and the
Penokean orogen. Geology 25, 415–418.
Reynolds, S.J., Lister, G.S., 1990. Folding of mylonitic zones in
Cordillerean metamorphic core complexes: evidence from near
the mylonitic front. Geology 18, 216–219.
Mountain Land. Geological Society of South Africa Special
Publication No. 9, pp. 21–29.
de Wit, M.J., Armstrong, R.A., Hart, R.J., Wilson, A.H., 1987.
Felsic igneous rocks within the 3.3–3.5 Ga Barberton greenstone
belt: high crustal level equivalents of the surrounding tonalitetrondhjemite terrain emplaced during thrusting. Tectonics 6,
529–549.
de Wit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., de Ronde,
C.E.J., Green, R.W.E., Tredoux, M., Peberdy, E., Hart, R.A.,
1992. Formation of an Archaean continent. Nature 357, 553–
562.
Dziggel, A., 2002. The Petrogenesis of ’Lower’ Onverwacht Group
Clastic Metasedmients and Related Metavolcanic Rocks in the
Southern Part of the Barberton Mountain Land, South Africa.
Ph.D. Thesis. University of the Witwatersrand, Johannesburg,
230 pp.
Dziggel, A., Stevens, G., Poujol, M., Anhaeusser, C.R., Armstrong,
R.A., 2002. Metamorphism of the Granite-Greenstone terrane
south of the Barberton greenstone belt, South Africa: An insight
into the tectono-thermal evolution of the lower portions of the
Onverwacht Group. Precambrian Res. 114, 221–247.
Ferry, J.M., 1984. A biotite isograd in south-central Main, USA:
mineral reactions, fluid transfer and heat transfer. J. Petrol. 25,
871–893.
Fripp, R.E.P., van Nierop, D.A., Callow, M.J., Lilly, P.A., du
Plessis, L.U., 1980. Deformation in part of the Archaean
Kaapvaal Craton, South Africa. Precambrian Res. 13, 241–251.
Graham, C.M., Powell, R., 1984. A garnet-hornblende
geothermometer: calibration, testing and application to the
Pelona schist, Southern California. J. Metamorph. Geol. 3, 13–
21.
Hamilton, W.B., 1998. Archaean magmatism and deformation were
not products of plate tectonics. Precambrian Res. 91, 143–179.
Heubeck, C., Lowe, D.R., 1994. Late syndepositional deformation
and detachment tectonics in the Barberton greenstone belt,
South Africa. Tectonics 13, 1514–1536.
Hill, E.J., 1994. Geometry and kinematics of shear zones formed
during continental extension in eastern Paua New Guinea. J.
Struct. Geol. 16, 1093–1105.
Holland, T.J.B., Powell, R., 1998. An internally consistent
thermodynamic data set for phases of geological interest. J.
Metamorph. Geol. 16, 309–344.
Jackson, M.P.A., Erikson, K.A., Harris, C.W., 1987. Early
Archaean foredeep sedimentation related to crustal shortening:
a reinterpretation of the Barberton sequence, southern Africa.
Tectonophysics 136, 197–221.
Kamo, S.L., Davis, D.W., 1994. Reassessment of Archean crustal
development in the Barberton Mountain Land, South Africa,
based on U–Pb dating. Tectonics 13, 167–192.
Kisters, A.F.M., Anhaeusser, C.R., 1995. Emplacement features
of Archaean TTG plutons along the southern margin of the
Barberton greenstone belt, South Africa. Precambrian Res. 75,
1–15.
Kohn, M.J., Spear, F.S., 1991. Two new barometers for garnet
amphibolites with applications to southeastern Vermont. Am.
Mineral. 75, 89–96.
Krabbendam, M., Dewey, J.F., 1998. Exhumation of UHP rocks
by transtension in the Western Gneiss Region, Scandinavian
378
(Ed.), Two Cratons and an Orogen—Excursion Guidebook
and Review Articles for a Field Workshop Through Selected
Archaean Terranes of Swaziland, South Africa and Zimbabwe.
IGCP project 280. Department of Geology, University of
Witwatersrand, Johannesburg, pp. 69–84.
Viljoen, M.J., Viljoen, R.P., 1969. An introduction to the geology
of the Barberton granite-greenstone terrain. Special Publication
of the Geological Society of South Africa 2, 9–28.
Visser, D.J.L., 1956. The geology of the Barberton area. Geological
Survey of South Africa Special Publication No. 15, 254 pp.
Williams, D.A.C., Furnell, R.G., 1979. A reassessment of part of
the Barberton area. Precambrian Res. 9, 325–347.
Xie, X., Byerly, G.R., Ferrel Jr., R.E., 1997. IIb trioctahedral
chlorite from the Barberton greenstone belt: crystal structure
and rock composition constraints with implications to
geothermometry. Contrib. Mineral. Petrol. 126, 275–291.
Ridley, J.R., Vearncombe, J.R., Jelsma, H.A., 1997. Relations
between Greenstone Belts and associated Granitoids. In: de Wit,
M.J., Ashwal, L.D. (Eds), Greenstone Belts. Oxford University
Press, Oxford, pp. 377–397.
Robb, L.J., Anhaeusser, C.R., 1983. Chemical and petrogenetic
characteristics of the Archaean tonalite-trondhjemite gneiss
plutons in the Barberton Mountain Land. Special Publication
of the Geological Society of South Africa 9, 103–116.
Stevens, G., Droop, G.T.R., Armstrong, R.A., Anhaeusser,
C.R., 2002. The Schapenburg schist belt: amphibolite-facies
metamorphism in Fig Tree Group metasediments as a
consequence of c. 3.23 Ga terrane accretion in the Barberton
greenstone belt, South Africa. S. Afr. J. Geol. 105, 271–
284.
Tomkinson, M.J., King, V.J., 1991. The tectonics of the
Barberton greenstone belt: an overview. In: Ashwal, L.D.
Precambrian Ophiolites and Related Rocks
Edited by Martin J. van Kranendonk, R. Hugh Smithies and Vickie C. Bennett
Developments in Precambrian Geology, Vol. 15 (K.C. Condie, Series Editor)
DOI: 10.1016/S0166-2635(07)15057-X
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Chapter 5.7: Metamorphism in the Barberton Granite Greenstone Terrain
34
36
2
35
37
1
2
METAMORPHISM IN THE BARBERTON GRANITE
GREENSTONE TERRAIN: A RECORD OF PALEOARCHEAN
ACCRETION
36
38
Chapter 5.7
3
GARY STEVENS AND JEAN-FRANCOIS MOYEN
37
39
1
11
38
40
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
10
9
8
7
6
5
4
12
39
41
41
Department of Geology, Stellenbosch University, Matieland, 7130, South Africa
13
40
42
43
42
41
40
39
38
37
36
35
34
33
32
41
43
Fig. 5.7-1. Geological map of the Barberton greenstone belt (modified after Anhaeusser et al.
(1981)). KaF: Kaap River Fault; KoF: Komatii fault; ISZ: Inyoni shear zone; IF: Inyoka–Saddleback
fault. The boxes refer to areas were the detailed metamorphic studies reviewed in this paper were
conducted: Western domain: (a) Stentor Pluton (Otto et al., 2005; Dziggel et al., 2006), (b) Schapenburg schist belt (Stevens et al., 2002). Eastern domain: (c) Tjakastad schist belt (Diener et al., 2005;
Diener et al., 2006), (d) Inyoni shear zone (Dziggel et al., 2002; Moyen et al., 2006), (e) Stolzburg
schist belt (Kisters et al., 2003), (f) Central Stolzburg terrane (Dziggel et al., 2002).
42
The Barberton Granite Greenstone Terrain (BGGT) has been interpreted to record an
accretionary orogeny during which at least two crustal terranes merged along a crustal
scale suture zone (de Ronde and de Wit, 1994; Lowe, 1994, 1999; de Ronde and Kamo,
2000). This orogeny has been deemed to be responsible for the main deformation event in
the Barberton Greenstone Belt (BGB) (D2), at ca. 3.21 Ga, which is well recorded in the
lower parts of the stratigraphy of the belt in, the Onverwacht and Fig Tree groups (Viljoen
and Viljoen, 1969c; Anhaeusser et al., 1981,1983; Lowe and Byerly, 1999; Lowe et al.,
1999). Terrane amalgamation was followed by the deposition of molasses of the Moodies
Group, which were themselves subsequently refolded during the late stages of orogeny.
In the nearby granitoids, ca. 3.23–3.21 Ga plutons are interpreted as resulting either from
arc-type magmatism, or from orogenic collapse (Moyen et al., this volume, and references
therein). Relatively high-grade metamorphism in the BGGT is confined to the granitoid
domains surrounding the belt and the Theespruit and Sandspruit Formations that form the
belt’s lower-most stratigraphy. The interior of the belt is typified by lower greenschist facies
metamorphism (Fig. 5.7-1) (Anhaeusser et al., 1981).
In the modern Earth, accretionary orogens involving collision between oceanic and continental plates are characterized by a particular pattern of regional metamorphic grade
distribution. In the lower plate (which is generally linked to a subducted oceanic plate),
high pressure and low to medium temperature metamorphism is developed (Chopin, 1984;
Bodinier et al., 1988; Ernst, 1988; Chopin et al., 1991; Nicollet et al., 1993; Spear, 1993;
Wang and Lindh, 1996), commonly reaching relatively high grades. In the upper plate,
lower grade metamorphism develops along typically warmer geotherms (Burg et al., 1984,
1989). This duality of metamorphic types has been recognized as one of the “hallmarks of
plate tectonics” and has been proposed as useful in determining the timing of the onset of
conventional plate tectonics (Brown, 2007). Thus far, clear evidence for this signature has
only been documented from the Proterozoic and Phanerozoic rock record (Brown, 2007).
In contrast, Archean metamorphic conditions are typically interpreted to reflect mostly
“hot” and uniform P-T conditions (Percival, 1994; Brown, 2007). Thus, Archean terrains
are regarded as lacking metamorphic evidence for collisional orogeny involving oceanic
rocks. Furthermore, the typical map pattern of gneissic domes surrounded by narrow, syn-
formal greenstone belts (“dome and keel patterns”) is regarded as contradictory with collision or collision-like processes (Chardon et al., 1996; Choukroune et al., 1997; Chardon
et al., 1998; Collins et al., 1998; Hamilton, 1998; Collins and Van Kranendonk, 1999; Van
Kranendonk et al., 2004; Bédard, 2006).
43
42
43
5.7-1. Evidence for Accretionary Orogeny in the BGGT
3
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1
40
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
20
41
43
Several new studies have recently been published on aspects of the metamorphic evolution of the BGGT and, in combination, provide particularly clear insights into the Archean
geodynamic processes that shaped the greenstone belt. In this chapter, we review the findings of these studies and show that two fundamentally important aspects emerge. Firstly,
that the higher-grade metamorphic margins to the belt are in faulted contact with the
lower-grade metamorphic interior, and that these zones are characterized by strong syndeformational isothermal decompression signatures, with peak metamorphic conditions
typically reflecting a minimum estimate (particularly for pressure). Secondly, there appear to be two fundamentally different metamorphic signatures in the amphibolite-facies
rocks associated with the belt. In the ca. 3.45 Ga and older granitoid-dominated terrane to
the south of the belt (Fig. 5.7-1), a relatively low-temperature, high-pressure metamorphic
signature is dominant. This contrasts with a significantly higher apparent geothermal gradient developed in the amphibolite-facies domains along granite-greenstone contacts on the
northern margin of the belt and within greenstone remnants in the far south of the BGGT.
The main body of the greenstone belt, although at lower metamorphic grades, also records
a signature of relatively high apparent geothermal gradient.
In addition to reviewing these metamorphic findings and their significance, this study
will propose a model for the development of the dome-and-keel pattern, within the framework of an orogenic process.
21
42
The general stratigraphy of the BGB appears to confirm the importance of tectonic
processes in the history of the belt. The stratigraphy of the BGB is subdivided into three
main groups, from bottom to top these are the Onverwacht, Fig Tree and Moodies Groups
(Viljoen and Viljoen, 1969c; Anhaeusser et al., 1981, 1983; Lowe and Byerly, 1999).
The 3.55–3.25 Ga Onverwacht Group predominantly consists of mafic/ultramafic lavas,
interstratified with cherts, rare clastic sedimentary rocks and felsic volcanic rocks. The
3.25–2.23 Ga Fig Tree Group is an association of felsic volcaniclastic rocks, together with
clastic and chemical [banded iron formation (BIF)] sedimentary rocks. The 3.22–3.21 Ga
Moodies Group is made of sandstone and conglomerates.
The Onverwacht, and, to some degree, the Fig Tree, Groups show different stratigraphies in the northwestern and southeastern parts of the BGB (Viljoen and Viljoen, 1969c;
Anhaeusser et al., 1981, 1983; de Wit et al., 1992; de Ronde and de Wit, 1994; Lowe, 1994;
Lowe and Byerly, 1999; Lowe et al., 1999; de Ronde and Kamo, 2000). In the west, the
Onverwacht Group is mostly 3.3–3.25 Ga, whereas it is much older in the eastern part of
the belt (3.55-3.3 Ga). Furthermore, the details of the stratigraphic sequences on both sides
cannot be correlated, confirming that the two parts of the belt evolved via a similar, yet independent history. The boundary between the two domains is tectonic and corresponds to
the Inyonka–Saddleback fault system, described below. This structure spans the length of
5.7-1.1. Stratigraphy
5.7-1. EVIDENCE FOR ACCRETIONARY OROGENY IN THE BGGT
43
42
43
4
the belt from the Stolzburg syncline near Badplaas in the south, to the northern extremity
at Kaapmuiden.
5.7-1.2. Tectonic History of the BGB
At least five major phases of deformation have been identified in the BGB (de Ronde and
de Wit, 1994; Lowe, 1999b; Lowe et al., 1999). Early D1 (ca. 3.45 Ga old) deformation is
occasionally preserved in lower Onverwacht Group rocks. However, the dominant tectonic
event recorded in these rocks occurred between 3.25 and 3.20 Ga. Four (or five) successive
deformation phases related to this event are identified. The first (D2a ) deformation occurred
during the deposition of the sedimentary and felsic volcanic rocks of the Fig Tree Group, at
3.25–3.23 Ga, probably associated with the development of a volcanic arc in what is now
the terrane to the west of the Inyoni–Inyoka fault system (discussed below). At ca. 3.23 Ga
(D2b ), a dominant period of deformation resulted from the accretion of the two terranes
along the Inyoni–Inyoka fault system.
The D2 accretion was immediately followed, at ca. 3.22–3.21 Ga, by the syn-tectonic
(D3 ) deposition of the sandstone and conglomerates of the Moodies Group in small and discontinuous, fault-bounded basins (Heubeck and Lowe, 1994a, 1994b). The D3 deformation
is at least in part extensional, with normal faulting in the BGB (upper crust) and core complex exhumation followed by diapiric rise of gneissic domes in the lower crust (surrounding
granitoids) (Kisters et al., 2003, 2004). This event corresponds to post-collisional collapse.
Late, ongoing compression resulted in strike-slip faulting and folding of the whole sequence, including the Moodies Group, during D4 and D5 deformation.
5.7-1.3. The Inyoka–Inyoni Fault System
Within the BGB, the main D2 structure is the “Inyoka–Saddleback fault”, which is developed approximately parallel to the northwestern edge of the belt (Lowe, 1994, 1999;
Lowe et al., 1999). This fault forms the boundary between the northwestern and southeastern facies of the Onverwacht Group. The fault system also contains several layered
mafic/ultramafic complexes (Anhaeusser, 2001), which may correspond to fragments of
oceanic crust trapped in a suture zone. On a larger scale, this zone corresponds to a geophysical boundary within the Kaapvaal craton that extends for several hundreds of kilometers along strike and separates two geophysically and geochronologically distinct terranes
(Poujol el al., 2003; de Wit et al., 1992; Poujol, this volume). The Inyoka–Saddleback fault
is made of a network of subvertical faults that were active during several of the later deformation events described above, leading to a complex history. It is interpreted to be a D2
thrust, that was steepened during subsequent (D3 –D5 ) deformation.
Further south in the granitoid dominated terrane, a ductile north–south trending shear
zone runs from the southern termination of the Stolzburg syncline towards the Schapenburg
schist belt, some 30 km further south. This zone, called the “Inyoni shear zone” (ISZ:
Kisters et al., 2004; Moyen et al., 2006), is a major structure in the granitoid terrane south
of the BGB; it separates the ca. 3.2 Ga Badplaas gneisses to the west, from the ca. 3.45 Ga
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.7-2. Metamorphic History of the Eastern Terrane
5
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
37
6
36
38
1
37
39
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
4
3
2
5
38
40
Stolzburg pluton in the east, mirroring the difference between the relatively young, western
“Kaap Valley” block and the older terranes (Songimvelo, etc.; Lowe, 1994) to the east of
the Inyoka–Saddleback fault. Thus, the ISZ is possibly a lower crustal equivalent of the
Inyoka–Saddleback fault system.
6
39
41
41
5.7-2.1. The Stolzburg Terrane
Amphibolite facies metamorphic domains have been investigated in detail in both the
Eastern and Western domains around the BGGT (Fig. 5.7-1). These potentially provide a
window into the lower or middle crust of different portions of the orogen.
5.7-2. METAMORPHIC HISTORY OF THE EASTERN TERRANE
40
42
5.7-2.1.1. Peak of metamorphism
Dziggel et al. (2002) documented two types of clastic metasedimentary rocks: a trough
cross-bedded, proximal meta-arkose and a planar bedded, possibly more distal, metasedimentary unit of relatively mafic geochemical affinity. The latter are characterized by
the peak-metamorphic mineral assemblage diopside + andesine + garnet + quartz. This
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
41
43
Fig. 5.7-2. Typical peak metamorphic textural relationships (left) and P-T estimates (right) for samples from the central Stolzburg terrane: (a) and (b) represent two examples of the post tectonic
peak metamorphic textures. On the P-T diagram; (c) BE1 and BE2 illustrate the peak metamorphic conditions as constrained by two of the samples studied by Dziggel et al. (2002). Schematic
andalusite-sillimanite-kyanite phase boundaries are included for reference.
42
One of the best studied high-grade regions in the BGGT is known as the “Stolzburg terrane”
(Kisters et al., 2003, 2004), which crops out to the south of the BGB, and corresponds to a
portion of the “Songimvelo block” of Lowe (1994). The Stolzburg terrane is comprised of
ca. 3.45 Ga trondhjemitic orthogneisses of the Stolzburg, Theespruit and other plutons. The
terrane contains greenstone material in the form of amphibolite-facies greenstone remnants
along the pluton margins, as well as amphibolite-facies Theespruit Formation rocks along
the southern margin of the BGB (Fig. 5.7-1). The greenstone remnants within the granitoid terrane have been interpreted to be part of the Sandspruit Formation of the Onverwach
Group (Anhaeusser et al., 1981, 1983; Dziggel et al., 2002) and consist of metamorphosed
mafic and ultramafic metavolcanic sequences, with minor metasedimentary units that comprise thin metachert and metamorphosed BIF interbanded with metamorphosed ultramafic
and mafic volcanic rocks. In addition to these typical lower Onverwacht Group lithologies,
this area also contains an up to 8 m-thick, metamorphosed clastic sedimentary unit, within
which are well-preserved primary sedimentary features, such as trough cross-bedding.
A minimum age of sediment deposition is indicated by a 3431 ± 11 Ma age of an intrusive trondhjemite gneiss (Dziggel et al., 2002). The youngest detrital zircons within the
metasedimentary rocks are 3521 Ma in age, indicating that the sedimentary protoliths were
deposited between ca. 3521 and 3431 Ma (Dziggel et al., 2002), and therefore are not
significantly older than the “overlying” Theespruit and Komatii Formations.
The Stolzburg terrane is bounded to the west by the ISZ, which separates it from
the 3.23–3.21 Ga Badplaas pluton, which therefore belongs to the Eastern domain. The
northern limit of the Stolzburg terrane is the Komati fault, which corresponds to a sharp
metamorphic break between the amphibolite-facies Stolzburg terrane and the greenschistfacies rocks of the main part of the BGB (Eastern domain: Kisters et al., 2003; Diener et
al., 2004).
Three recent studies are relevant to the metamorphism of this terrane: Dziggel et al.
(2002), who studied the metamorphism of rare clastic metasedimentary rocks within greenstone remnants along the southern margin of the Stolzburg pluton; Kisters et al. (2003),
who studied the tectonometamorphic history of the northern boundary of the Stolzburg
pluton; and Diener et al. (2005), who investigated the tectonometamorphic history of the
Tjakastad schist belt (areas c, e, and f on Fig. 5.7-1).
43
42
43
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
7
25
40
42
Fig. 5.7-2. (Continued.)
41
43
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
42
assemblage (and garnet in particular) is extensively replaced by retrograde epidote. Peakmetamorphic mineral assemblages of magnesio–hornblende + andesine + quartz, and
quartz + ferrosilite + magnetite + grunerite have been recorded from adjacent amphibolites and interlayered BIF units, respectively. In these rocks, retrogression is marked by
actinolitic rims around peak metamorphic magnesio–hornblende cores in the metamafic
rocks, and by a second generation of grunerite that occurs as fibrous aggregates rimming
orthopyroxene in the iron formation. The peak metamorphic textures are typically post
tectonic and are texturally mature and well equilibrated. Peak pressure-temperature (PT)
estimates, using a variety of geothermometers and barometers, for the peak-metamorphic
mineral assemblages in all these rock types vary between 650–700 ◦ C and 8–11 kbar
(Fig. 5.7-2). As suggested by the texturally well-equilibrated nature of the assemblages,
no evidence of the prograde path is preserved. Dziggel et al. (2002) interpreted the relatively high pressures and low temperatures of peak metamorphism to reflect a tectonic
setting comparable to modern continent–continent collisional settings, and suggested that
the Stolzburg terrane represents an exhumed mid- to lower-crustal terrane that formed a
‘basement’ to the BGB at ca. 3230 Ma.
43
42
43
8
5.7-2.1.2. Contacts with the greenstone belt
The deformed and metamorphosed margins of the Stolzburg terrane in the north, where
it abuts the lower grade greenstone belt, have been studied in two separate areas. Kisters
et al. (2003) conducted detailed mapping of the contacts between the supracrustal and
gneiss domains along the southern margin of the greenstone belt. They documented an
approximately 1 km wide deformation zone that corresponds with the position of the heterogeneous and mélange-like rocks of the Theespruit Formation, within which two main
strain regimes can be distinguished (Fig. 5.7-3). Amphibolite-facies rocks at, and below,
the granite–greenstone contacts are characterized by rodded gneisses and strongly lineated
amphibolite-facies mylonites. Lineations developed in the BGB either side of the Stolzburg
syncline are brought into parallelism by unfolding around the inclined fold axis of the syncline, suggesting extension prior to folding. When rotated into a subhorizontal orientation,
the bulk constrictional deformation at these lower structural levels records the originally
vertical shortening and horizontal, NE–SW directed stretching of the mid-crustal rocks.
The prolate coaxial fabrics are overprinted by greenschist-facies mylonites at higher structural levels that cut progressively deeper into the underlying high-grade basement rocks.
These mylonites developed during non-coaxial strain and kinematic indicators consistently
point to a top-to-the-NE sense of movement of the greenstone sequence with respect to
the lower structural levels. This relationship between bulk coaxial NE–SW stretching of
mid-crustal basement rocks and non-coaxial, top-to-the-NE shearing along retrograde mylonites at upper crustal levels is consistent with an extensional orogenic collapse of the belt
and the concomitant exhumation of deeper crustal levels.
The dominant peak metamorphic assemblage within preserved amphibolite-facies domains throughout the study area is hornblende + plagioclase + sphene + quartz. Other
locally developed assemblages are: garnet + hornblende + plagioclase + sphene + quartz,
and garnet + plagioclase + hornblende + calcite + biotite + epidote + quartz in metamafic rocks; and garnet + biotite + muscovite + quartz in a single metapelitic layer. All
the garnet-bearing assemblages are confined to specific narrow layers developed parallel
to the compositional banding of the rocks (S0 ). In all cases, retrogression is associated
with the development of later shear fabrics (S1 in retrograde mylonites) that postdate the
peak-metamorphic porphyroblasts.
Kisters et al. (2003) interpreted these features to suggest a primary bulk-compositional
control (Fe/Fe+Mg ratios and the presence of carbonate) on the distribution of the garnetbearing peak-metamorphic assemblages, and that these assemblages are probably metamorphic grade equivalents of the predominant peak assemblage in the amphibolites. Peak
P-T conditions were constrained using the assemblages garnet + plagioclase + hornblende
+ biotite + quartz, and garnet + plagioclase + hornblende + biotite + quartz + epidote +
calcite, which yielded P-T estimates of 491±40 ◦ C and 5.5±0.9 kbar, and 492±40 ◦ C and
6.3 ± 1.5 kbar, respectively. Retrogression is marked by the development of actinolite +
epidote + chlorite + quartz assemblages in the metamafic rocks and muscovite + chlorite
+ quartz in the metapelitic layer. These conditions are at lower grades than those defined
by Dziggel et al. (2002), but are developed along a similarly low apparent geothermal
gradient.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
9
35
37
36
38
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Fig. 5.7-4. Typical peak metamorphic textural relationships (left) and P–T estimates (right) for samples from the Tjakastad area (Diener et al.,
2005, 2006): (a) Illustrates two generations of syntectonic garnet development; (b) illustrates a typically deformed plagioclase porphyroblast;
(c) illustrates the P-T conditions of metamorphism calculated using assemblages from the Tjarkastad schist belt. The sample numbers in
(c) correspond to those used by Diener et al. (2005).
43
10
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
aid: 15057 pii: S0166-2635(07)15057-X docsubty: REV
F:dpg15025.tex; VTEX/JOL p. 10
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
9
11
10
8
7
6
5
4
3
2
1
43
42
41
40
39
37
Fig. 5.7-3. Schematic cross-sections across granite–greenstone contacts from the Western and Eastern domains of the BGGT. (a) The low to high grade transition in the Stentor pluton area in the
Western domain (after Dziggel et al., 2006). (b) The northern boundary of the Stolzburg terrane
against the Eastern domain (after Kisters et al., 2003).
38
39
40
41
42
43
5.7-3. Metamorphism in the Western Domain
11
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1
40
42
41
40
39
38
37
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
35
41
43
Diener et al. (2004) investigated the tectonometamorphic history of the Tjakastad schist
belt (Fig. 5.7-1), which contains remnants of the Theespruit Formation that predominantly
includes amphibolites, felsic volcanoclastic rocks, and minor aluminous metasedimentary
rocks. The metamafic and metasedimentary rocks record an identical deformational history to the rocks studied by Kisters et al. (2003), some 5 to 10 km to the northwest. Both
the peak metamorphic and retrograde assemblages are syntectonic with fabrics developed
during exhumation, illustrating the initiation of detachment at deep crustal levels and elevated temperatures. In contrast with the rocks studied by Kisters et al. (2003), however, the
rocks investigated by Diener et al., (2004) provided a better record of the retrograde path.
Within the metamafic rocks, more aluminous layers are characterized by the peak metamorphic assemblage garnet + epidote + hornblende + plagioclase + quartz. Within the
aluminous metasedimentary unit, an equivalent peak metamorphic assemblage is defined
by garnet + staurolite + biotite + chlorite + plagioclase + quartz. These assemblages
produce calculated P-T estimates of 7.0 ± 1.2 kbar and 537 ± 45 ◦ C and, 7.7 ± 0.9 kbar and
563 ± 14 ◦ C, respectively (Fig. 5.7-4). In these rocks, the peak metamorphic assemblages
are syntectonic, with peak metamorphic porphyroblasts (e.g., staurolite) recrystallised and
deformed within the exhumation fabric (Fig. 5.7-4). Within rare low-strain domains in the
garnet-bearing amphibolite, retrograde mineral assemblages pseudomorph peak metamorphic garnet. In these sites, a new generation of garnet is developed within the assemblage
garnet + chlorite + muscovite + plagioclase + quartz. Calculated P-T estimates from
these sites yield conditions of 3.8 ± 1.3 kbar and 543 ± 20 ◦ C, indicating near isothermal
decompression (Fig. 5.7-4). This is consistent with the presence of staurolite as part of
the peak and retrograde assemblages, with the modeled staurolite stability field in relevant
compositions being confined to a narrow temperature range of between 580–650 ◦ C over a
pressure range between 10–3 kbar. These calculated P-T conditions are also consistent with
the occurrence of sillimanite replacing kyanite within the staurolite-bearing rocks (Diener
et al., 2004).
Geochronological constraints, combined with the depths of burial, indicate that exhumation of the high-grade rocks occurred at rates of 2–5 mm/a. This is similar to the
exhumation rates of crustal rocks in younger compressional orogenic environments, and
when coupled with the low apparent geothermal gradients of ca. 20 ◦ C/km, led Diener et
al. (2004) to suggest that the crust was cold and rigid enough to allow tectonic stacking,
crustal overthickening and an overall rheological response very similar to that displayed
by modern, doubly-thickened continental crust.
36
42
The metamorphic history of the Western domain is less well understood than the
Stolzberg terrane, as fewer studies have been conducted and these are more widespread,
making the relationships between the study areas less obvious. Two studies are relevant
to this discussion: the study by Dziggel et al. (2006), who investigated the tectonometamorphic history of the northern contact of the BGB, where it is in contact with the Stentor
5.7-3. METAMORPHISM IN THE WESTERN DOMAIN
43
42
43
12
pluton [area (a) in Fig. 5.7-1]; the study by Stevens et al. (2002), who investigated the
metamorphic history of the Schapenburg schist belt [area (b) in Fig. 5.7-1]. This study area
lies along the southern extension of the ISZ, which is believed to anastomose around the
Schapenburg schist belt. This belt is included in the Western domain on account of it displaying a similar apparent geothermal gradient to that documented by Dziggel et al. (2006).
An important difference between the Western and Eastern domains is that the Eastern domain contains an abundance of granitoid intrusions (Badplaas, Nelshoogte and Kaapvalley)
that are essentially syntectonic with the ca 3.23 Ga deformation.
5.7-3.1. Schapenburg Schist Belt
The Schapenburg schist belt is one of several large (approximately 3 × 12 km) greenstone
remnants exposed in the granitoid-dominated terrane to the south of the BGB and is unique
in that it contains a well-developed metasedimentary sequence in addition to the typical
mafic-ultramafic volcanic rocks (Anhaeusser, 1983). Stevens et al. (2002) conducted an
investigation of the metamorphic history of the belt, which is summarized below.
The metasedimentary sequence consists of two distinctly different units. A metatuffaceous unit, essentially of granitoid composition, but containing both minor agglomerate layers and, within low strain domains, well preserved cross-bedding and graded
bedding in the southwestern portion of the belt. This unit underlies a rhythmically banded
unit of metagreywacke that consists of approximately 10 cm-thick units of formerly clayrich rock that grade into 1 to 2 cm thick quartz-rich layers. On the basis of both the graded
bedding and trough cross-bedding in the underlying meta-tuffaceous unit, the metasedimentary succession can be shown to young to the east. This succession is overlain by
Onverwacht Group rocks.
Detrital zircons within the metasedimentary rocks have ages as young as 3240 ± 4 Ma
and thus are correlated with the Fig Tree Group in the central portions of the BGB some
60 km to the north, where they are metamorphosed to lower greenschist facies grades.
The Schapenburg schist belt metasedimentary rocks are relatively K2 O-poor and are
commonly characterized by the peak metamorphic assemblage garnet + cordierite +
gedrite + biotite + quartz ± plagioclase. Other assemblages are garnet + cummingtonite
+ biotite + quartz, cordierite + biotite + sillimanite + quartz and cordierite + biotite
+ anthophyllite. In all cases, the post-tectonic peak assemblages are texturally very well
equilibrated (Fig. 5.7-5) and the predominantly almandine garnets from all rock types show
almost flat zonation patterns for Fe, Mg, Mn and Ca. Consequently, there appears to be no
preserved record of the prograde path.
Analysis of peak metamorphic conditions using FeO-MgO-Al2 O3 -SiO2 -H2 O FMASH
reaction relations, as well as a variety of geothermometers and barometers, constrained the
peak metamorphic pressure-temperature conditions to 640 ± 40 ◦ C and 4.8 ± 1.0 kbar. The
maximum age of metamorphism was defined by the 3231 ± 5 Ma age of a syntectonic
tonalite intrusion into the central portion of the schist belt. In combination with the age
of the youngest detrital zircons in the metasedimentary rocks, this age demonstrates that
sedimentation, burial to mid-crustal depths (∼18 km), and equilibration under amphibolite
facies conditions were achieved in a time span of between 10–20 Ma.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
24
34
36
14
25
35
37
13
26
36
38
5.7-3. Metamorphism in the Western Domain
27
37
39
31
32
33
34
35
36
37
38
39
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
28
38
40
40
41
29
39
41
30
40
42
Fig. 5.7-5. Typical peak metamorphic textural relationships (left) and P-T estimates (right) for samples from the Schapenburg schist belt (Stevens et al., 2002): (a) and (b) illustrates the typically post
tectonic character of the peak metamorphic minerals (garnet in (a) and garnet and orthoamphibole
in (b); (c) P-T diagram illustrating the calculated conditions of peak metamorphism. The sample
numbers correspond to those used by Stevens et al. (2002).
41
43
5.7-3.2. Stentor Pluton Area
42
Dziggel et al. (2006) showed that the granitoid–greenstone contact along the northern margin of the BGB is characterized by a shear zone that separates the generally low-grade,
greenschist-facies greenstone belt from mid-crustal basement gneisses. The supracrustal
rocks in the hangingwall of this contact are metamorphosed to upper greenschist facies,
whereas similar rocks and granitoid gneisses in the footwall are metamorphosed to amphibolite facies. Within the amphibolite facies domain, metamafic rocks are characterized
by the assemblages hornblende + plagioclase + quartz; hornblende + plagioclase +
clinopyroxene + quartz and hornblende + plagioclase + garnet + clinopyroxene + quartz.
Aluminous schists from this domain contain the peak metamorphic assemblage garnet +
muscovite + sillimanite + biotite + quartz.
Calculated P-T estimates on these assemblages constrain the peak P-T conditions of
metamorphism to between 600 and 700 ◦ C and 5 ± 1 kbar (Fig. 5.7-6). This corresponds
to an elevated geothermal gradient of ∼30–40 ◦ C/km. The peak metamorphic minerals in
this area are syntectonic with fabrics that are interpreted to have formed during exhumation
of the high grade rocks at ca 3.23 Ga. Retrograde assemblages that form through the replacement of peak metamorphic clinopyroxene and plagioclase in the metamafic rocks by
coronitic epidote + quartz and actinolite + quartz symplectites yield retrograde P-T conditions of 500–650 ◦ C and 1–3 kbar. This indicates that exhumation and decompression
commenced under amphibolite facies conditions (as indicated by the synkinematic growth
of peak metamorphic minerals during extensional shearing), followed by near-isobaric
43
42
43
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
21
34
36
16
22
35
37
15
23
36
38
5.7-4. Inyoni Shear Zone
24
37
39
27
28
29
30
31
32
33
34
35
36
37
5.7-4. INYONI SHEAR ZONE
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
38
40
40
41
39
41
26
40
42
Fig. 5.7-6. Typical peak metamorphic textural relationships (left) and P-T estimates (right) for samples from the Stentor pluton area (Dziggel et al., 2006): (a) and (b) illustrate the mineral assemblages
studied by Dziggel et al. (2006); (c) P-T diagram, with sample numbers used by Dziggel et al. (2006).
38
41
43
cooling to temperatures below 500 ◦ C. The last stages of exhumation are characterized
by solid state doming of the footwall gneisses and strain localization in contact-parallel
greenschist-facies mylonites that overprint the decompressed basement rocks.
The southern margin of the Stentor pluton area is bounded by the Kaap River and Lily
faults (Fig. 5.7-1). These correspond to a major metamorphic break, from 6–8 kbar in the
amphibolitic domain, to nearly unmetamorphosed supracrustal rocks in the BGB immediately south of the faults (Otto et al., 2005; Dziggel et al., 2006).
39
42
The Inyoni shear zone (ISZ) is a complex structure extending in a southwesterly direction from the termination of the Stolzburg syncline into the granitoid dominated terrane to
the south (area (d) in Fig. 5.7-1). It forms the boundary between the Stolzburg terrane to the
east and the Badplaas pluton to the west. Both Dziggel et al. (2002) and Moyen et al. (2006)
have investigated the metamorphic history of the ISZ. The shear zone contains a diverse
assemblage of greenstone remnants, mostly typical lower Onverwacht Group interlayered
metamafic and meta-ultramafic units, with occasional minor BIF horizons, but some clastic metasedimentary rock also occur (Dziggel et al., 2002). The greenstone fragments are
enclosed within TTG orthogneisses, components of which were intruded syntectonically
during, or close to, the peak of metamorphism.
Structures in the Inyoni shear zone are complex, and result from the interference of:
(1) east–west shortening, resulting in the formation of a predominantly vertical foliation,
with symmetrical folds and the development of a crenulation cleavage at all scales
(from the map pattern to hand specimen); and
(2) vertical extrusion of the Stolzburg terrane, causing the development of a syn-melt vertical lineation, and folds with vertical axes.
Evidence for earlier structures has also been described, in the form of rootless isoclinal
folds in some of the supracrustal remnants.
43
42
43
43
42
41
40
Table 5.7-1.
39
38
37
36
35
32
31
30
Retrograde
minerals
33
34
27
26
25
24
29
19
18
17
16
P
15
14
10
T
12
9
7
S.D.
(T)
8
4
3
2
1
23
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
41
40
39
38
37
36
34
32
31
30
25
24
23
22
21
Notes Method
17
P
14
13
S.D.
(P)
27
26
8
7
6
5
4
6
3
2
1
18
Ref
9
S.D.
(T)
Dziggel et al., 2006
”
”
10
T
190
40
30
11
753
630
675
20
19
18
16
15
12
12
7
”
1.2
”
[5]
[5]
30
Hbl-Pl (ed-tr)
Hbl-Pl (ed-ri)
640
30
113
113
690
159
[5]
575–700
THERMOCALC (av. 5.7
PT)
[5]
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
”
”
Stevens et al., 2002
”
3–5.7
74
79
625–725
633
654
5–6
1.2
”
5.4
<5.3 kbar
at 650 ◦ C
4.8
1.1
475–650
159
159
159
peak
78c
Pseudosection
modelling
(THERMOCALC)
Pseudosection
modelling
(THERMOCALC)
retro
THERMOCALC
(av. PT)
Petrogenetic grid
(THERMOCALC)
THERMOCALC
(av. PT)
SKG53
G8b
ca. 3.5
Grt-Bi (Ganguly and
Saxena, 1994)
Grt-Bi (Hackler and
Wood, 1989)
Pseudosection
modelling
(THERMOCALC)
Sample
number
28
Ctd
Ser
Retrograde
minerals
33
Grt + Crd + Ged
+ Bi + Pl + Qtz
± Cumm
Mu + Bi + Qtz
+ And
Grt + Mu +
Sili-Bi-Qtz
Peak
assemblage
35
Schapenburg
Metaturbidite
Felsic schist
(greenschist
facies)
Felsic schist
(amphibolite
facies)
Table 5.7-1. (Continued)
42
”
”
”
43
”
Cloete, 1991, 1999
Ref.
5
ca. 520
ca. 350
Xye et al., 1997
11
11
6
6
Ref.
ca. 320
S.D.
(T)
S.D.
(P)
13
Al substitution in
chlorite
T
Chl, Amp and PI ca. 4
isopleths +
Hbl-Pl
Chl, Amp and Pl ca. 2.5
isopleths +
Hbl-Pl
Method
20
retro
peak
Sample Notes
number
28
Peak
assemblage
A) Within the BGB proper
Komatii formation
Mafic/ultramafic Chl + Amp + Pl
pillows
Onverwacht & Fig Tree, center of the belt
Diverse, mostly mafic to
Chl-Trem/Act +
intermediate lavas
Qtz ± Ser ± Cc
Greenschist
facies
assemblage
S.D.
(P)
22
21
23
22
21
Method
P
Sample Notes
number
Retrograde
minerals
35
Peak
assemblage
595
16
15
12
670
620
[600]
”
”
”
0.7
[700]
50
40
42
20
4.6
2.9
165a
3.2
Ep + Zo + Act
B) Amphibolite facies – Western domain
Stentor pluton
Metabasites
[5]
[5]
6
Hbl + Pg + Qtz
± Cpx ± Grt
154
625
600
6.5
[5]
[5]
165a
165a
154
THERMOCALC
(av. PT)
Grt-Cpx
Grt-Hbl
THERMOCALC
(av. P)
THERMOCALC
(av. P)
Hbl-Pl (ed-tr)
Hbl-Pl (ed-ri)
154
154
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
20
19
18
17
14
13
10
9
8
7
5
4
3
2
1
17
43
42
41
40
39
38
37
36
35
34
33
32
29
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
9
11
10
8
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
31
30
25
24
Notes
20
19
18
Method
29
27
26
23
22
21
17
16
11
S.D.
(P)
9
T
8
7
S.D.
(T)
5
Ref.
10
10
6
6
4
3
2
1
42
41
40
39
38
37
36
35
34
32
31
Retrograde
minerals
33
29
28
27
26
25
Sample Notes
number
15
P
10
T
5
Ref.
20
19
18
13
12
9
8
7
4
3
2
1
20
17
16
14
11
11
6
6
”
”
S.D.
(T)
668–753
[700]
630–706
[10]
601–652
S.D.
(P)
Hbl-Pl (ed-ri)
[10]
8.2–12.1
660–706
Method
21
BE1
Hbl-Pl (ed-tr)
Cpx-Pl-Qtz (Ellis,
1980)
[10]
[10]
24
26
25
23
24
22
23
22
12
7
”
Hbl-Pl (ed-tr)
Hbl-Pl (ed-ri)
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
”
636–694
”
Opx (Witt-Eiskschen [10]
and Seek, 1991)
Gru(n+1) + Act SL1-5
+ Mag(n+1)
641–749
”
Dziggel et al. 2002
625–756
Qtz + Mag +
Gru + Opx +
Hbl
[10]
[700]
639–767
[10]
”
”
[10]
8.7–9.9
[700]
[700]
Cpx + Pl + Qtz Ep ± Ser ± Act SL1-6
+ Grt ± Hbl
8.1–11.5
9.0–11.0
Grt-Cpx (Ellis and
Green, 1979)
Grt-Cpx (Ganguly,
1979)
Grt-Cpx (Powell,
1985)
Grt-Cpx-Pl-Qtz
(Powell and Holland,
1988)
Grt-Cpx-Pl-Qtz
(Eckert et al., 1991)
Cpx-Pl-Qtz
Hbl + Pl + Qtz Ep + Ser + Act BE2
+ Cpx
30
Clastic
metasediments
Inyoni shear zone
Iron formation
Amphibolite
Central Stolzburg – Greenstone remnant BE
Clastic
Hbl + Pl + Cpx Ser
metasediments + Qtz
Peak
assemblage
43
”
7
”
11
569 18
Diener et al., 2005
Diener et al., 2005
543 20
”
556 19
7.2 1
556 59
7.9 1.1
3.8 1.3
537 45
”
1.2
1.6
563 14
7
Kisters et al., 2003
7
491 40
”
”
”
5.5 0.9
557 37
458 35
569 42
1.3
8
5.1 0.9
6.1 2.7
7.7 0.9
P
12
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
13
62105F
THERMOCALC
(av. PT)
peak
max, for XH2 O
=1
retro
62601C
Tj18
62107
Tj3
61406
An1
min, for XH2 O
= 0.3
XH2 O = 0.5
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
Sample
number
28
Retrograde
minerals
32
Peak
assemblage
C) Amphibolite facies – Eastern domain
Grt(n+1) +
Chl(n+1) +
Pl(n+1) + Mu
Act + Ep + Chl
+ Qtz
Grt + Ep + Pl + Chl
Hbl + Qtz
Tjakastad schist belt
Metasediment
Grt + St + Bi +
Chl + Pl + Qtz
Metabasites
Stolzburg arm (exhumation)
Amphibolites
Hbl + Pg + Qtz
+ Sph ± Grt
PB3
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
15
17
16
14
15
14
13
12
9
8
5
4
3
2
1
19
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
21
20
19
18
17
16
15
14
13
9
10
8
5
4
3
2
1
43
42
41
40
39
38
37
36
35
32
31
Retrograde
minerals
33
26
25
24
23
22
21
20
19
18
17
16
12
11
S.D.
(P)
9
T
5
Ref
14
13
10
8
7
6
6
4
3
2
1
30
36
35
31
30
29
28
23
22
Notes
26
25
24
24
21
20
16
15
14
10
8
7
5
3
2
T
68
74
S.D.
(T)
”
”
”
Ref.
1
653
585
690
725
58
39
50
50
”
”
”
”
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
7.5
1.7
1.0
22
580
156
2
S.D.
(P)
646
1
2.8
827
3
P
3.1
”
”
5
13.5
2.2
125
”
4
15.7
758
95
”
7
9.5
1.8
764
117
6
9.4
1.3
740
108
8
8.8
1.4
830
10
6.7
1
1.3
9
8.7
7.9
”
10.1
”
8
11
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. P)
Hbl-Pl (avg of 2
reactions)
THERMOCALC
(av. P)
Hbl-Pl (avg of 2
reactions)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
Method
17
Grt breakdown
18
”
Grt core
Grt rim
”
”
”
”
Matrix
”
Matrix
591a
Grt breakdown
”
INY25
INY21
Sample
number
27
(different sites
for each sample)
Retrograde
minerals
32
Peak
assemblage
37
”
dpg15 v.2007/05/23 Prn:31/05/2007; 11:37
”
7
668 66
38
”
690 99
39
”
2.4
40
604 102
”
”
2.7
41
599 100
623 60
14.9
42
3.8
13
3.7
695 100
16.4
43
”
10
[700]
”
”
11
8.9–11.0
[700]
[700]
”
”
8.0–9.9
8.7–10.4
[700]
”
S.D.
(T)
9.9–11.9
[700]
[700]
10.3–11.6
Moyen et al., 2006
4.1
14.1
2.2
”
3.3
604 61
14.0
2.6
13.0
12.8
11.7
13.1
534 92
7.9–10.7
P
15
Method
27
SL1-8
Sample Notes
number
SL1-8a
INY131 Grt growth site
”
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
THERMOCALC
(av. PT)
Grt-Cpx-Pl-Qtz
1988)
Grt-Cpx-Pl-Qtz
Cpx-Pl-Qtz
Grt-Cpx-Pl-Qtz
1988)
Grt-Cpx-Pl-Qtz
Cpx-Pl-Qtz
28
Hbl + Ep + Pl
+ Grt ± Cpx
Peak
assemblage
Garnetamphibolites
Ep(n+1) +
Hbl(n+1) +
Pl(n+1)
Chl-Trem/Act
”
”
”
”
43
42
41
40
39
38
37
36
34
35
34
33
32
31
29
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
12
9
8
5
4
3
2
1
21
43
42
41
40
39
38
37
34
36
35
33
34
33
32
31
30
29
28
19
13
12
27
26
25
23
22
21
20
19
18
17
16
15
14
13
12
9
11
6
4
1
5.7-5. Discussion and Conclusions
23
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
37
24
36
38
1
37
39
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
38
40
40
43
42
41
40
39
38
37
36
35
34
39
41
41
Fig. 5.7-7. Typical metamorphic textural associations (left) and P-T estimates (right) for samples
from the Inyoni shear zone (Moyen et al., 2006): (a) Illustrates the small garnet crystals developed
in conjunction with albitic plagioclase during the breakdown of sodic hornblende; (b) illustrates an
intergrowth of garnet and clinopyroxene; (c) P-T diagram, with sample numbers after Moyen et al.
(2006).
38
40
42
Both greenschist and amphibolite facies remnants have been described, possibly indicating the imbrication of rocks with diverse metamorphic histories. However, most of the
remnants are dominated by metamafic rocks and appear to have been metamorphosed to
amphibolite facies grades. The dominant foliation is defined by hornblende in the metamafic rocks, which is cut by syntectonic tonalitic veins with an age of 3229 ± 5 Ma
(Dziggel et al., 2006). Metamorphic titanite that formed in association with epidote through
retrograde replacement of garnet and plagioclase has an age of 3229 ± 9 Ma (Dziggel et
al., 2006).
Dziggel et al. (2002) focused on metasedimentary rocks within the ISZ and produced
P-T estimates of amphibolite facies peak metamorphic conditions very similar to those
described for the Stolzburg terrane, at 600–700 ◦ C and 8–11 kbar. No information on the
prograde history of the rocks could be determined due to the well equilibrated nature of
the peak metamorphic assemblages.
In contrast, Moyen et al. (2006) examined the metamorphic record within metamafic
samples and produced information on both the prograde and retrograde P-T evolution of
this zone. Textural evidence of the prograde metamorphic evolution is recorded in garnetbearing low-strain domains, such as the cores of certain rootless isoclinal folds, where
core-to-rim growth-zoned garnet occurs that have low-temperature mineral inclusions contained within their cores. In these sites, garnet can be shown to have grown simultaneously with albitic plagioclase, as evidenced by euhedral garnets surrounded by plagioclase
(Fig. 5.7-7) and albitic inclusions within garnets, sometimes with negative garnet forms.
In the same domains, clinopyroxene and quartz are also sometimes intergrown with garnet (Fig. 5.7-7). This assemblage appears to have formed at the expense of a relatively
sodic amphibole (Fe-edenite, up to 1.1 sodium atoms per formula unit), partially reequilibrated relicts of which are found within albitic moats around the garnets. These commonly
occur as several small, separate relic crystals that are in crystallographic continuity, indicating the original presence of substantially larger crystals. Calculated P-T estimates for
this assemblage in a number of sites range from 600–650 ◦ C and 12–15 kbar. Garnet in
samples from higher-strain domains generally shows partial replacement by symplectitic
coronas of epidote + Fe-tschermakite + quartz symplectite. Calculated P-T estimates from
these assemblages produce retrograde conditions of 580–650 ◦ C at 8–10 kbar. The estimated metamorphic conditions constrained by these decompression structures correspond
well with peak metamorphic estimates from the nearby clastic sedimentary intercalations
within the metavolcanic sequence. Locally, in both the high- and low-strain domains,
greenschist-facies chlorite + epidote + actinolite retrogression overprints the amphibolitefacies assemblages.
39
41
43
5.7-5. DISCUSSION AND CONCLUSIONS
42
The data presented above support several general observations on the nature of the ca.
3.23 Ga metamorphic event in the BGGT:
(1) The existence of two different thermal regimes in the deep crust of the BGGT. Midto lower-crustal rocks from the Western domain generally record apparent geothermal
gradients as low as 18–20 ◦ C km−1 . Similar rocks from the Eastern domain record
apparent geothermal gradients of 30–40 ◦ C km−1 .
43
42
43
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
21
34
36
26
22
35
37
25
23
36
38
24
37
39
29
30
31
32
33
34
35
36
37
38
39
43
42
41
40
39
38
37
36
35
34
31
30
29
28
27
26
25
25
38
40
40
41
27
39
41
26
28
40
42
Fig. 5.7-8. Compilation diagram of P-T estimates of the studies discussed in this paper. Strong evidence for decompression exists in the samples from the Inyoni shear zone, the Tjakastad schist belt
and the Stentor pluton. The rocks of the Eastern domain clearly underwent a peak of metamorphism
in the kyanite stability field, potentially recording heating during exhumation from greater depths
than the recorded pressures indicate. Peak metamorphic conditions in the Western domain were in
the sillimanite stability field.
41
43
Fig. 5.7-9. Proposed geodynamic model for the ca. 3.2 Ga accretionary orogen in the BGGT. All
cartoons are at approximately the same scale, looking towards the (present-day) northeast; the front
section of each block corresponds to a NW–SE cross-section. In each cartoon, the active plutonism
is in black, while the already emplaced rocks are grey. Plutons: B: Badplaas, N: Nelshoogte, KV:
Kaapvalley, S: Stolzburg, Ts: Theespruit. Structures: IF: Inyoka–Saddleback fault, ISZ: Inyoni shear
zone. Cartoons are modified from (Moyen et al., 2006). Circled letters (A, B, C, D) in the figures
correspond to the Theespruit Formation of the Tjakastad schist belt (Diener, 2005), the ISZ samples (Moyen et al., 2006), the Schapenburg schist belt (Stevens, 2002), and the Stentor pluton area
(Dziggel et al., 2006), respectively. The P-T evolution of points A, B and C during the assembly and
collapse phases of the orogen are illustrated in the P-T diagrams presented below the second and
third cartoons.
42
(2) The granite-greenstone margins in both domains are defined by the presence of amphibolite facies supra-crustal rocks and gneissose granitoids, as part of the deep crustal
section. In both cases, a substantial pressure transition of >5 kbar (ca. 15 km) can
be documented across the sheared contacts, over just a few kilometers laterally. This
transition occurs in a zone of high-strain rocks (up to mylonites) that record a normal
sense of movement with the low-grade greenstone belt being down-thrown relative to
the surrounding amphibolite-facies gneisses. In essence, these zones define the cuspate granite-greenstone contacts of the “dome and keel” pattern. Peak metamorphism
in these areas is syntectonic with the exhumation process, which is continuous as the
margin of the uplifted block evolved into greenschist facies conditions.
(3) In those parts of both the Eastern and Western domains, peak metamorphic conditions
away from the greenstone belt are post-tectonic. This indicates coherent behavior of
the exhumed deep crust, in that the mappable domains discussed here, such as the
Stolzburg terrane, represent largely intact deep crustal sections that were exhumed
along discrete shear zones along the granite-greenstone contacts. This lack of penetra-
43
42
43
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
27
42
43
28
tive post-peak metamorphic deformation internal to the terranes appears inconsistent
with the diapiric rise of plastically deforming domes.
(4) The peak P–T estimates for the ISZ, as well as the mélange-like character of the zone
(Moyen et al., 2006), confirm this zone as a terrane boundary and the possible trace of
the subduction zone that closed to allow crustal collision. The pressures reported for
this zone (P = 12 to 15 kbar) are, at present, the highest crustal pressures reported for
meso-Archean rocks, and correspond to by far the lowest known apparent geothermal
gradients (12 ◦ C km−1 ) in the Archean rock record. In the modern Earth, the only
process capable of producing crustal rock evolution through this P–T domain occurs
within subduction zones.
5.7-5.1. The Case for 3.2 Ga Cold Crust and Horizontal Tectonics
The case for cool continental crust in the BGGT prior to 3.23 Ga is convincing. The rocks
of the Stolzburg terrane represent an approximately 400 km2 domain of rocks that were
buried to depths of 35–40 km. Internally to this domain, peak metamorphic equilibration
occurred, in rocks that were not undergoing deformation, to record an amphibolite facies
apparent geothermal gradient no higher than those recorded by younger metamorphic rocks
from ocean-continent collision zones. This occurred simultaneously with syntectonic peak
metamorphism in the terrane margins, where deformation was driven by the exhumation
of the high-grade portions of the thickened crust. The presence of crustal rocks recording
pressures of 12–15 kbar and an apparent geothermal gradient of 12 ◦ C km−1 , in the setting
of a shear zone containing both metasedimentary and metamafic rocks at variable grades
of peak metamorphism, is used to suggest that this zone marks the prior existence of a
subduction zone.
The abundance of synkinematic trondhjemites in the shear zone is likely to be the result
of decompression melting of amphibolites at deeper levels during exhumation. The presence of these melts is possibly important to understanding the documented metamorphic
signature. High-strain fabrics confined to synkinematic trondhjemites point to strain localization in the melts, which, in turn, is likely to assist the buoyancy- or extrusion-related
exhumation of the rocks. The advective heat transfer associated with the intrusion of these
synkinematic magmas also contributes to the syn- to late-collisional heat budget of the
collisional belt that acted to partially destroy the evidence for the earlier high-pressure,
low-temperature metamorphism. This possibly holds the key to understanding the very
high geothermal gradients recorded by the high-grade rocks of the western domain following uplift (50–60 ◦ C km−1 ), as much of the crust that constitutes the western domain
comprises syntectonic magmatic rocks.
5.7-5.2. Are the BGGT Domes Core Complexes?
Sections of the Pilbara Craton in Western Australia and the BGGT show numerous
regional-scale similarities. For this discussion the most notable of these are the typical
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
29
dome-and-keel geometries between TTG domes and greenstone synforms, and the localized occurrence of high-pressure, low- to medium-temperature metamorphism of the
supracrustal sequences (Collins et al., 1998; Van Kranendonk et al., 2002; Van Kranendonk, 2004a). Importantly, despite these similarities, completely different models have
arisen for the evolution of the Archean crust in these two areas. Tectonic models proposed to account for the high-grade metamorphism of greenstone sequences in the Pilbara
Craton critically hinge on a high rate of heat production in the Mesoarchean crust, that is
assumed to be in excess of twice typical modern day rates (e.g., Sandiford and McLaren,
2002) and has produced significant weakening of the crust (e.g., Marshak, 1999). Thus,
the high-grade metamorphism, special thermal regime and widespread constrictional-type
strains recorded in the Pilbara supracrustal succession, are interpreted to indicate the gravitational sinking and burial of denser, mainly mafic and ultramafic greenstone rocks along
the flanks of, and between, buoyant and rising TTG diapirs, a model known as partial
convective overturn of the crust (e.g., Collins et al., 1998; Van Kranendonk et al., 2002,
2004a).
In contrast, the 3.23 Ga structural evolution of the Stolzburg terrane in the BGGT has
been interpreted to be the result of core-complex like exhumation of the lower crust, probably in a post-collision setting (Kisters et al., 2003). The following points appear to argue
strongly against partial convective overturn of the crust in the BGGT of the type proposed
for the Pilbara Craton.
(1) The southern contact of the BGB with the Stolzburg terrane is marked by mainly prolate fabrics. These are exhumation fabrics and they are not related to the burial or
sinking of the supracrustal sequence.
(2) The highest pressures (8–11 kbar by Dziggel et al. (2002), and 12–15 kbar by Moyen
et al. (2006)) are documented from the southern TTG terrain and not in the greenstone sequences. Indeed, the felsic plutonic rocks contain metamorphic assemblages
recording significantly higher pressures than the flanking greenstones.
(3) In addition, the low apparent geothermal gradient in the exhumed basement to the
south of the BGGT appears inconsistent with an essentially thermally driven process.
Despite these important differences, there are some similarities in the processes proposed for the two areas. After initiation of exhumation of the granitoid domains along
the extensional detachment in the southern BGGT, there is a transition from extensional,
to buoyancy driven rise, and the final emplacement of the gneissic “domes” may well be
aided by the buoyancy contrast between the gneisses and the mafic greenstones. In the
later stage, ascent of the coherent basement blocks causes the development of a predominantly linear fabric, with vertical stretching lineations along their margins. Unlike the
scenarios proposed in other, similar, dome-and-keel terranes, the ascent here occurs after
an initial stage of extensional collapse, and affects not single magma batches (plutons), nor
migmatitic complexes, but essentially chunks of solid, composite continental crust made
of several well-identified plutons and surrounding volcanosedimentary sequence (within
which lithological relationships, including intrusion relationships between the plutons and
the supracrustal rocks ca. 200 Ma prior to the orogenic history, are often well-preserved).
This is mostly a solid-state process, although syntectonic intrusion into the high-strain mar-
30
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
1
41
43
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
42
gins is common and possibly important in achieving the significant vertical displacement
recorded by the magnitude of the metamorphic pressure differences across the margins.
This late evolution and steepening of bounding shear zones to close to vertical is not classically known from modern core complexes, but seems to correspond to a unique Archean
process that is essentially driven by the buoyancy contracts between the mafic/ultramafic
lower sections of the greenstone belt stratigraphy and the granitoid middle and lower crust.
It may be possible that in Archean orogens (at least in the BGGT), crustal thickening
followed by orogenic collapse quickly evolves into buoyancy driven, near-vertical emplacement of the lower-crustal domains as a result of the higher density contrast between
the heavy upper crust (dominated by mafic/ultramafic rocks) and the felsic lower crust
(TTG gneisses), resulting in a density inversion and an unstable density stratification. Such
a situation is not commonly attained in modern orogens, where the upper crust is made of
lighter gneisses or sediments, and the lower crust of dense eclogites or granulites. It might
be tempting to also propose a higher Archean heat production, causing a generally softer
lithosphere, and facilitating bulk diapiric rise of the crust. In the case of the BGGT, this
does not appear to fit either the relatively low-temperature, high-pressure metamorphic signature of the Western domain, or the strain localization patterns associated with the uplift
of this rather rigid crustal block.
43
Vol 442|3 August 2006|doi:10.1038/nature04972
LETTERS
Record of mid-Archaean subduction from
metamorphism in the Barberton terrain, South Africa
559
expose sections through different crustal levels of the ,3.23-Gyr
collisional orogen.
This study presents the results of a metamorphic analysis of rare
mineral assemblages found in supracrustal remnants from within a
prominent shear zone within this gneiss terrain, the Inyoni shear
zone, which is probably the mid- to lower crustal expression of the
suture that accommodated the mid-Archaean terrane accretion. The
Inyoni shear zone is an up to 3 km wide, north-trending subvertical
belt of banded, often migmatitic gneisses (Fig. 1). It extends southwards, to the kilometre-scale amphibolite-facies Schapenburg Schist
belt. In the west, it is intruded by syntectonic bodies of coarsegrained, leucocratic trondhjemites to granodiorites. Some of these
D2 bodies yielded ages of 3.229 ^ 0.005 Gyr (ref. 12) and
3.231 ^ 0.005 Gyr (ref. 13), constraining the timing of the deformation. Towards the east, the zone is bounded by relatively homogeneous and lower-strain gneisses of the high-grade Stolzburg
terrane. Although heterogeneous strain and high degrees of fabric
transposition make it difficult to establish the overall kinematics and
strain within the shear zone, the scarcity of non-coaxial fabrics and
the fabric geometry point to a dominantly bulk flattening strain
associated with a component of vertical extrusion of the rocks. The
gneisses of the Inyoni shear zone contain metre- to kilometre-scale,
variably deformed and metamorphosed metavolcanic and subordinate metasedimentary remnants. Lithological differences between
mappable packages, both in the Inyoni shear zone proper, and in the
Schapenburg Schist belt, together with contrasting pressure–
temperature (P–T) conditions (8–11 kbar and 650–700 8C (ref. 14)
in the North; ,5 kbar and 630 8C in Schapenburg) suggest that this
composite gneiss belt is a tectonic melange, juxtaposing rocks from
diverse crustal depths intruded by largely synkinematic granitoids. A
northern metavolcanic package (Fig. 1) consists predominantly of
layered, epidote- and hornblende-dominated amphibolites. Garnet
occurs within specific, relatively iron-rich horizons and the metamorphic history of this zone can best be understood by focusing on
these pressure-sensitive garnet-bearing assemblages.
Prograde metamorphic evolution is recorded in low-strain
domains, such as the cores of rootless isoclinal folds, where garnet
grew simultaneously with albitic plagioclase, as evidenced by euhedral garnets surrounded by plagioclase (Fig. 2a) or albitic inclusions
within garnets, sometimes with negative garnet forms. Clinopyroxene
and quartz are sometimes intergrown with garnet (Fig. 2b). This
assemblage formed at the expense of a relatively sodic amphibole
(Fe-edenite, up to 1.1 sodium atoms per formula unit), and epidote,
partially reequilibrated relicts of which are found in crystallographic
continuity within albitic moats around the garnets. Qualitatively,
garnet–clinopyroxene–quartz assemblages are known to form at
relatively high pressures15. In coexisting garnet–plagioclase pairs,
Ca is preferentially partitioned into garnet over plagioclase as
pressure increases15; thus, relatively calcic garnets coexist with sodic
Jean-François Moyen1, Gary Stevens1 & Alexander Kisters1
Although plate tectonics is the central geological process of the
modern Earth, its form and existence during the Archaean era
(4.0–2.5 Gyr ago) are disputed1,2. The existence of subduction
during this time is particularly controversial because characteristic subduction-related mineral assemblages, typically documenting apparent geothermal gradients of 15 8C km21 or less3,
have not yet been recorded from in situ Archaean rocks (the lowest
recorded apparent geothermal gradients 4 are greater than
25 8C km21). Despite this absence from the rock record, low
Archaean geothermal gradients are suggested by eclogitic nodules
in kimberlites5,6 and circumstantial evidence for subduction
processes, including possible accretion-related structures2, has
been reported in Archaean terrains. The lack of spatially and
temporally well-constrained high-pressure, low-temperature
metamorphism continues, however, to cast doubt on the relevance
of subduction-driven tectonics during the first 1.5 Gyr of the
Earth’s history7. Here we report garnet–albite-bearing mineral
assemblages that record pressures of 1.2–1.5 GPa at temperatures
of 600–650 8C from supracrustal amphibolites from the midArchaean Barberton granitoid-greenstone terrain. These conditions
point to apparent geothermal gradients of 12–15 8C—similar to
those found in recent subduction zones—that coincided with the
main phase of terrane accretion in the structurally overlying
Barberton greenstone belt8. These high-pressure, low-temperature
conditions represent metamorphic evidence for cold and strong
lithosphere, as well as subduction-driven tectonic processes, during
the evolution of the early Earth.
Recent studies have highlighted the composite nature of the earlyto mid-Archaean Barberton granitoid-greenstone terrain and have
demonstrated that the deep crustal levels (30–40 km) are exposed in
structurally bounded domains in the granitoid-gneiss terrain to the
south of the shallow-crustal greenstone belt9,10. Sedimentological,
structural and geochronological differences indicate that the belt is
made up of a northern and a southern terrane that are separated by
the central, NE–SW trending Saddleback–Inyoka fault. The amalgamation of these two proposed island-arc terranes occurred during
the main, D2 phase of collisional tectonics at ,3.23 Gyr ago,
probably in an arc-trench setting8. The surrounding, granitoidgreenstone terrain is made up of (1) an amphibolite-facies greenstone component, (2) mainly gneissic trondhjemitic plutonic rocks
,3.55–3.45 Gyr old, (3) syntectonic (D2) trondhjemites and tonalites
,3.23–3.22 Gyr old11. The older trondhejmites and the associated
greenstone remnants form an extensive, relatively high-grade domain
(the Stolzburg terrane, Fig. 1), that was metamorphosed to pressures
of up to 8–11 kbar (refs 9, 10). Significantly, peak metamorphic
conditions in the high-grade gneiss terrain were attained during the
D2 phase of tectonism, coeval with the accretion of the two island-arc
terranes in the shallow-crustal greenstone belt. In other words, the
supracrustal greenstone belt and the deep-crustal gneiss terrain
© 2006 Nature Publishing Group
1
Department of Geology, University of Stellenbosch, South Africa Private Bag X-1, Matieland 7602, South Africa.
NATURE|Vol 442|3 August 2006
background represents the felsic, trondhjemitic and tonalitic (TTG)
gneisses. Dotted lines display foliation trends. BIF, banded iron formation.
Inset, stereograms (Schmidt, lower hemisphere) for the central domains of
the mapped area. The star in b denotes the location of the studied samples,
which were derived from different amphibolite bodies.
LETTERS
Figure 1 | Location of the Inyoni shear zone and studied samples in the
southern Barberton terrane14. a, Regional geological map. Ages are from
refs 11 and 21. ISZ, Inyoni shear zone. The box indicates the location of the
detailed map in b. The darker-grey domain and the areas marked with
smaller crosses correspond to the high-grade ,3.45-Gyr Stolzburg block.
b, Detailed geological map of the northern Inyoni shear zone. The white
plagioclase. e, Garnet surrounded by epidote–quartz symplectites (INY25).
f, Core-to-rim zoned garnet, with a low-temperature core with albite,
epidote and amphibole inclusions rimmed by a higher-temperature rim in
equilibrium with more calcic plagioclase and amphibole; a ring of quartz
inclusions bounds the core (dashed line) (INY21). Amp, amphibole (Amp1,
early Na-rich amphibole); Pl, plagioclase (Ab, albitic plagioclase); Gt, garnet;
Cpx, clinopyroxene; Ep, epidote; Qz, quartz.
Figure 2 | Metamorphic textures associated with garnet growth or
breakdown (plane-polarized light). a, Small euhedral garnets rimmed by
albitic plagioclase, coalescing into large, poekilitic grains (INY131).
b, Garnet–clinopyroxene intergrowth associated with garnet formation; the
large garnet–clinopyroxene grain is rimmed by albite, not seen at this
magnification (INY115). c, Breakdown of sodic amphibole within the albitic
moats rimming garnets (INY131). d, Albitic plagioclase with euhedral grains
of garnet and epidote, evidence of the breakdown of an earlier, more calcic
560
plagioclase at the highest pressures of plagioclase-and-garnet coexistence (Fig. 3). The prograde garnet generation documented in
this study is indeed calcic (35–40% grossular) and coexists with almost
pure sodic endmember albitic plagioclase (An3–10). The garnet-in
reaction is steeply orientated in P–Tspace in the area where it intersects
the high-pressure plagioclase phase boundary. The magnesium
number Mg# (Mg/Mg þ Fe) in garnet scales inversely with temperature away from this phase boundary. The low Mg# (10–15) of the
prograde garnet in these rock compositions (Mg# < 50) argues for
LETTERS
Figure 3 | THERMOCALC P–T estimates for garnet growth and breakdown
sites in the studied samples. Filled ellipses are the average for 11 (garnet
growth, solid lines) and 9 (garnet breakdown, dashed lines) sites; empty
ellipses are examples of individual calculations, each corresponding to one
single reaction site using the compositions of minerals in textural
equilibrium. Different mineral assemblages were used (for example, garnet–
clinopyroxene–plagioclase–quartz, and garnet–amphibole–plagioclase–
epidote–quartz, for the garnet growth sites); they all produce
indistinguishable P–T estimates within error. Imprecision associated with
the plagioclase activity model, especially for low-An contents22, creates a
relative error on An activity of 5–15% in garnet growth sites (low-An
plagioclase) and 2–5% in garnet breakdown sites (intermediate plagioclase
composition). An additional source of error is that inherent to the
calibration of the garnet–clinopyroxene–plagioclase–quartz barometer22.
These uncertainties are integrated within the THERMOCALC P–T
estimates23,24. Despite the significant error ellipses, the uncertainty on the
apparent geotherm is much lower, owing to the positive slope of the
reactions used in P–T estimates.
low-temperature garnet formation (Fig. 3). This approach is confirmed by estimates using THERMOCALC16 thermobarometry (analytical techniques and representative mineral estimates are given in
Supplementary Tables 1 and 2, respectively), consistently pointing to
conditions of formation of 12–15 kbar and 600–650 8C (Fig. 3) for
the garnet-bearing assemblage. Garnet from samples in the highstrain domains (samples INY 25, 591a) generally shows retrograde
textures (Fig. 2e). Garnet breakdown conditions, recorded by
epidote þ Fe-tschermakite þ quartz symplectite coronas around
561
Komatii fault. Plutons are: Ts, Theespruit; KV, Kaap valley; N, Nelshoogte;
B, Badplaas; S, Stolzburg. d, e, The P–T of points A, B and C during the
assembly and collapse phases of the orogen are shown. The two hatched
blocks marked ‘metasediments of ISZ’ in e are from ref. 9.
Figure 4 | Geodynamic sketch summarizing the inferred tectonic evolution
of the southern Barberton terrane, and the associated metamorphic
evolution. a–c, Circles lettered A, B and C correspond to the Theespruit
(Ts) formation of the Tjakastad schist belt10, the ISZ samples from this study,
and the Schapenburg greenstone belt13, respectively. IF, Inyoka fault; KF,
LETTERS
9. Dziggel, A., Stevens, G., Poujol, M., Anhaeusser, C. R. & Armstrong, R. A.
Metamorphism of the granite–-greenstone terrane south of the Barberton
greenstone belt, South Africa: an insight into the tectono-thermal evolution of the
‘lower’ portions of the Onverwacht group. Precambr. Res. 114, 221–-247 (2002).
10. Diener, J., Stevens, G., Kisters, A. F. M. & Poujol, M. Metamorphism and
exhumation of the basal parts of the Barberton greenstone belt, South Africa:
constraining the rates of mid-Archaean tectonism. Precambr. Res. 143, 87–-112
(2005).
11. Kamo, S. L. & Davis, D. W. Reassessment of Archean crustal development in
the Barberton mountain land, South-Africa, based on U-Pb dating. Tectonics 13,
167–-192 (1994).
12. Dziggel, A., Armstrong, R. A., Stevens, G. & Nasdala, L. Growth of zircon and
titanite during metamorphism in the granitoid-gneiss terrain south of the
Barberton greenstone belt, South Africa. Mineral. Mag. 69, 1021–-1038 (2006).
13. Stevens, G., Droop, G. T. R., Armstrong, R. A. & Anhaeusser, C. R. Amphibolitefacies metamorphism in the Schapenburg schist belt: a record of the midcrustal response to ,3.23 Ga terrane accretion in the Barberton greenstone
belt. S. Afr. J. Geol. 105, 271–-284 (2002).
14. Kisters, A. F. M., Stevens, G., Dziggel, A. & Armstrong, R. A. Extensional
detachment faulting and core-complex formation in the southern Barberton
granite–-greenstone terrain, South Africa: evidence for a 3.2 Ga orogenic
collapse. Precambr. Res. 127, 355–-378 (2003).
15. Kohn, M. J. & Spear, F. S. Empirical calibration of geobarometers for the
assemblage garnet þ hornblende þ plagioclase þ quartz. Am. Mineral. 74,
77–-84 (1989).
16. Holland, T. J. B. & Powell, R. An internally consistent thermodynamic
dataset for phases of petrological interest. J. Metamorph. Geol. 16, 309–-343
(1998).
17. Diener, J. F. A., Stevens, G. & Kisters, A. F. M. High-pressure low-temperature
metamorphism in the southern Barberton granitoid-greenstone terrain, South
Africa: a record of overthickening and collapse of mid-Archean continental
crust. In Archean Geodynamics And Environments (eds Benn, K., Mareschal, J.-C.
& Condie, K.) 239-354 (AGU Geophysical Monograph Series Vol. 164, AGU,
Washington, 2005).
18. Chemenda, A. I., Mattauer, M. & Bokun, A. N. Continental subduction and a
mechanism for exhumatin of high-pressure metamorphic rocks: new modelling
and field data from Oman. Earth Planet. Sci. Lett. 143, 173–-182 (1996).
19. Nicollet, C. & Leyreloup, A. Pétrologie des niveaux trondjhémitiques de haute
pression associés aux éclogites et amphibolites des complexes leptyno
amphiboliques du Massif Central français. Can. J. Earth Sci. 15, 695–-707 (1978).
20. Bodinier, J. L., Burg, J.-P., Leyreloup, A. & Vidal, H. Reliques d’un bassin
d’arriere arc subducté puis obducté dans la région de Marvejols (Massif
Central). Bull. Soc. Geol. Fr. 8, 20–-34 (1988).
21. de Ronde, C. E. J. & Kamo, S. L. An Archaean arc-arc collisional event: a shortlived (ca 3 Myr) episode, Weltvreden area, Barberton greenstone belt, South
Africa. J. Afr. Earth Sci. 30, 219–-248 (2000).
22. Spear, F. S. Metamorphic Phase Equilibria and Pressure-Temperature-Time Paths
535 (Mineralogical Society of America, Washington, 1993).
23. Powell, R., Holland, T. J. B. & Worley, B. Calculating phase diagrams involving
solid solutions via non-linear equations, with examples using THERMOCALC.
J. Metamorph. Geol. 16, 577–-588 (1998).
24. Holland, T. J. B. & Blundy, J. Non-ideal interactions in calcic amphiboles and
their bearing on amphibole-plagioclase thermometry. Contrib. Mineral. Petrol.
116, 433–-447 (1994).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements J.-F.M.’s post-doctoral stay at Stellenbosch university is
funded by the South African National Research Fundation (NRF) and by a
bursary from the Department of Geology, Stellenbosch University. Running costs
were provided by the NRF. We thank G. Droop and J. Bédard for reviews of
earlier versions of this manuscript.
Author Contributions J.-F.M. and G.S. contributed equally to the metamorphic
and petrologic analysis. All authors contributed to the interpretation of these
results within the Barberton geodynamic framework.
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare no competing
financial interests. Correspondence and requests for materials should be
addressed to J.-F.M. ([email protected]) or G.S. ([email protected]).
Hamilton, W. B. Archean magmatism and deformation were not products of
plate tectonics. Precambr. Res. 91, 143–-179 (1998).
de Wit, M. J. On Archaean granites, greenstones, cratons and tectonics: does
the evidence demand a verdict? Precambr. Res. 91, 181–-226 (1998).
Ernst, W. G. Tectonic history of subduction zones inferred from retrograde
blueschist P-T paths. Geology 16, 1081–-1084 (1988).
Riciputi, L. R., Valley, J. W. & McGregor, V. R. Conditions of Archean
granulite-facies metamorphism in the Gothåb-Fiskenaesset region, southern
West Greenland. J. Metamorph. Geol. 8, 171–-190 (1990).
Rollinson, H. Eclogite xenoliths in West African kimberlites as residues from
Archaean granitoid crust formation. Nature 389, 173–-176 (1997).
Ireland, T. R., Rudnick, R. L. & Spetius, Z. Trace elements in diamond inclusions
from eclogites reveal link to Archaean granites. Earth Planet. Sci. Lett. 121,
199–-213 (1994).
Bjørnerud, M. G. & Austrheim, H. Inhibited eclogite formation: the key to the
rapid growth of strong and buoyant Archean continental crust. Geology 32,
765–-768 (2004).
De Wit, M. J. et al. Formation of an Archaean continent. Nature 357, 553–-562
(1992).
Received 17 March; accepted 12 June 2006.
the garnets, correspond to (THERMOCALC) temperatures of 580–
650 8C at 8–10 kbar. This set of metamorphic conditions is consistent
with the position of the (negatively sloped) garnet phase boundary in
this part of the P–T space (Fig. 3); the estimated metamorphic
conditions from these decompression structures corresponds well
with peak metamorphic estimates from the nearby clastic sedimentary intercalations within the metavolcanic sequence9. The peak
pressure P–T estimates are at present the highest crustal pressures
reported for Archaean rocks, and correspond to by far the lowest
known apparent geothermal gradients (,12 8C km21) in the
Archaean rock record. In the modern Earth, the only process capable
of producing crustal rock evolution through this P–T domain occurs
within subduction zones.
The Inyoni shear zone is the structurally and lithologically composite western boundary of the structurally coherent, high-pressure,
low-temperature Stolzburg granitoid-gneiss terrane. The presence of
rocks with a high-pressure history consistent with a subduction
origin in this zone suggests that this may conceivably represent the
suture along which the high-grade continental Stolzburg terrane was
rapidly buried to depths of at least 35–40 km (Fig. 4) (refs 9, 17). We
suggest that the mélange-like character of the shear zone is the result
of the structural imbrication of deeply buried slivers during the
buoyancy-assisted return flow between or close to the downgoing
slab and the overriding plate18. The abundance of synkinematic
trondhjemites in the shear zone is likely to be the result of decompression melting of amphibolites during the retrograde exhumation
path. The presence of these melts is possibly important to understanding the documented metamorphic signature. High-strain fabrics
confined to synkinematic trondhjemites point to strain localization
into the melts, which, in turn, is likely to assist the buoyancy- or
extrusion-related exhumation of the rocks. The advective heat
transfer associated with the intrusion of these synkinematic magmas
also contributes to the syn- to late-collisional heat budget of the
collisional belt that acted to partially destroy the evidence for the
earlier high-pressure, low-temperature metamorphism. In many
respects, this is similar to high-pressure amphibolites from more
recent subduction–collision belts, which often occur as partially
retrogressed boudins within migmatites19,20. Our findings of highpressure, low-temperature metamorphic mineral assemblages from an
active margin setting strongly suggest that lithospheric subduction was
functioning as early as 3.2 Gyr ago before present.
1.
2.
3.
4.
5.
6.
7.
8.
562
Metamorphism and exhumation of the basal parts of the Barberton
greenstone belt, South Africa: Constraining the rates of
Mesoarchaean tectonism
a
Received 11 August 2004; received in revised form 8 September 2005; accepted 3 October 2005
b
Department of Geology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa
Department of Earth Sciences, Memorial University of Newfoundland,
300 Prince Philip Drive, St. John’s, Nﬂd, Canada A1B 3X5
Johann F.A. Diener a,∗ , Gary Stevens a , Alexander F.M. Kisters a , Marc Poujol b
Abstract
The Paleo- to Mesoarchaean Barberton granitoid-greenstone terrain of South Africa consists of two main components: the
low-grade metamorphic supracrustal greenstone sequence of the Barberton greenstone belt in the north and a high-grade metamorphic granitoid-gneiss terrain to the south. The boundary between the two different domains corresponds to the highly tectonized,
amphibolite-facies rocks of the Theespruit Formation that occur along the margins of the greenstone belt. These rocks record high-P,
low-T peak metamorphic conditions of 7.4 ± 1.0 kbar and 560 ± 20 ◦ C that are very similar to estimates from other areas of the
high-grade terrain and were attained during the main phase of terrain accretion in the greenstone belt at 3230 Ma. In contrast, the
greenstone sequence ca. 4 km to the north only records low greenschist-facies metamorphism, indicating that a metamorphic break of
ca. 18 km exists between the high-grade terrain and the greenstone belt. The main phase of deformation in the Theespruit Formation
was initiated under peak metamorphic conditions and continued during retrogression. Retrograde P–T estimates and mineral reactions indicate that retrogression involved near-isothermal decompression of ca. 4 kbar prior to cooling into the greenschist-facies,
suggesting that the fabric in these rocks is an exhumation fabric that accommodated the juxtaposition of the high-grade terrain against
the greenstone belt. Geochronological constraints, combined with the depths of burial indicate that exhumation of the high-grade
rocks occurred at rates of 2–5 mm/a and are comparable to the exhumation rates of crustal rocks in younger orogenic environments.
The extremely low apparent geothermal gradients of ca. 20 ◦ C/km that are recorded in the high-grade terrain are inconsistent with
models of a hotter and weaker crustal environment in the Archaean. Rather, the depths of burial and structural integrity of this
terrain suggest that the Mesoarchaean crust was cold and rigid enough to allow tectonic stacking and crustal overthickening and
had a rheology similar to modern continental crust.
Almost all of the higher grade metamorphic rocks
developed in Archaean granitoid-greenstone terrains
exhibit penetrative fabrics that are indicative of regional
metamorphic processes. Whereas similarly deformed
rocks from younger metamorphic belts have provided
1. Introduction
Keywords: Archaean tectonics; Metamorphism; Exhumation rates; Barberton greenstone belt
∗ Corresponding author. Present address: School of Earth Sciences,
University of Melbourne, Vic. 3010, Australia. Tel.: +61 3 83444996;
fax: +61 3 83447761.
E-mail address: [email protected] (J.F.A. Diener).
doi:10.1016/j.precamres.2005.10.001
88
The lower part of the BGB stratigraphy consists
of ultramafic to mafic lavas of the ca. 3550–3300 Ma
Onverwacht Group, which are overlain by clastic marine
sediments and felsic volcanics of the 3260–3225 Ma
Fig Tree Group. The stratigraphy is completed by the
continentally derived coarse-clastic sediments of the
3225–3215 Ma Moodies Group that unconformly overly
lithologies of both the Onverwacht and Fig Tree Groups
(Viljoen and Viljoen, 1969a,b; SACS, 1980). Along the
2. Regional geology
during their history. This has been confirmed by a recent
investigation of an area where these contrasting metamorphic grades are developed in association with the
margin of the Stolzburg Pluton (Fig. 1; Kisters et al.,
2003). This study has proposed the presence of an extensional detachment to account for the post-peak metamorphic juxtaposition of the two different crustal domains,
and has also suggested that amphibolite-facies portions
of the greenstone sequence exposed along the margins of
the Theespruit Pluton (Fig. 1) are likely to constitute part
of the high-grade granitoid-gneiss terrain. Thus, the juxtaposition of the granitoid-gneiss terrain against the BGB
is accounted for in an exhumation model that is akin to
core-complex formation in younger orogenic belts (e.g.
Davis, 1980; Lister and Davis, 1989). Prior studies have
viewed the higher grade marginal portions of the BGB as
the metamorphic sole of an obducted ophiolite (De Wit et
al., 1983, 1987; Armstrong et al., 1990; De Ronde and De
Wit, 1994), or the consequence of contact metamorphism
associated with shearing during the diapiric intrusion of
the TTG plutons (Anhaeusser, 1984). Despite the obvious potential of a metamorphic investigation to provide
clarity on the details of the exhumation process, a suitably detailed study has not yet been conducted.
This study aims to contribute to this developing understanding of the tectonic development of the Barberton
granitoid-greenstone terrain by investigating the tectonometamorphic history of the high-grade rocks in the
Tjakastad schist belt (TSB) and the areas around the
Theespruit Pluton (Fig. 1). This area straddles the metamorphic break between the low-grade greenstones and
high-grade gneiss terrain, and also coincides with the
occurrence of the basal detachment identified by Kisters
et al. (2003), as well as terrain amalgamation structures
and sheared pluton boundaries regarded as important by
earlier workers. Detailed knowledge of the timing, nature
and duration of metamorphism in these rocks has the
potential to provide clarity on the nature and rates of
one of the earliest recognized tectonic episodes in Earth
history.
information indispensable to the development of our
understanding of modern orogenic and tectonic processes (e.g. Miyashiro, 1961; Ernst, 1973, 1975, 1988;
Chopin, 1984; Smith, 1984), comparatively few studies have attempted to do the same for these significantly
older terrains (e.g. Williams and Currie, 1993). In particular, the potential of metamorphic studies to provide clarity on Paleo- to Mesoarchaean geodynamic issues has
remained largely untapped, perhaps primarily due to the
lower potential of the ultramafic to mafic to I-type granitoid crust that dominates granitoid-greenstone terrains to
accurately record metamorphic change (e.g. Will et al.,
1990). The resultant lack of a well-constrained metamorphic framework has hampered the development of
a geotectonic framework for most granitoid-greenstone
terrains. Consequently, little consensus exists as to the
applicability of lateral tectonic processes to Archaean
crustal development, the timing of the onset of ‘conventional’ tectonics within Earth history as well as the nature
of Archaean tectonic processes (e.g. Davies, 1992, 1995;
Condie, 1994; Hamilton, 1998; De Wit, 1998; Kusky and
Polat, 1999; Marshak, 1999).
The rocks of the Paleo- to Mesoarchaean Barberton
granitoid-greenstone terrain (e.g. Viljoen and Viljoen,
1969a,b; Anhaeusser, 1973, 1984; De Wit, 1982; De
Ronde and De Wit, 1994; Lowe, 1994) are unusual for
Archaean rocks in being typified by an emergent coherent metamorphic framework that relies largely on the
occurrence of rare aluminous units with a high sensitivity for P–T change. It has been established that the rocks
of the supracrustal greenstone sequence in the Barberton greenstone belt (BGB) have generally be subjected to
only low metamorphic grades (Xie et al., 1997; Cloete,
1999). Greenschist- to sub-greenschist-facies grades are
typical, and pressure is constrained to ≤4 kbar in the
highest grade rocks. The only exception to this is the
marginal portions of the greenstone sequence where
higher grades are recorded in contact with the bounding granitoid plutons. Here, as within the granitoidgneiss terrain to the south of the BGB, amphibolite-facies
conditions of regional metamorphism have been documented that contrast strongly with metamorphic conditions in the central parts of the belt (Stevens et al., 2002;
Dziggel et al., 2002; Kisters et al., 2003). Peak metamorphic conditions within this terrain vary; however, close to
the contact with the BGB, peak metamorphic pressures
have been constrained to between 8 and 11 kbar (Dziggel
et al., 2002). This suggests a substantial metamorphic
break between the BGB and the granitoid-gneiss terrain
to the south, and that these two components have experienced contrasting tectono-metamorphic histories and
may have been tectonically juxtaposed at some point
89
Fig. 1. Geological map and stratigraphic column of the southern part of the Barberton granitoid-greenstone terrain (modified after Anhaeusser et
al., 1981; Kröner et al., 1996). Cited ages are crystallization ages of the granitoid plutons and are from (a) Kröner et al. (1991); (b) Kamo and Davis
(1994) and (c) Kröner et al. (1996). P–T estimates of metamorphism are from (1) Xie et al. (1997); (2) Cloete (1999); (3) Dziggel et al. (2002) and
(4) Kisters et al. (2003). The locations of the Komati Fault (De Wit et al., 1983, 1987; Armstrong et al., 1990; De Ronde and De Wit, 1994) and
the proposed extensional detachment in the Stolzburg schist belt (Kisters et al., 2003) are also shown. The extent of the current study area in the
Tjakastad schist belt is indicated by the box.
90
Most of the lithologies in the TSB exhibit strongly
developed, subvertical tectonic fabrics and primary bedding (S0 ) features are only preserved in low-strain
domains. In general, the contacts between lithological units are of a tectonic, rather than of stratigraphic
nature and different litho-tectonic units are truncated
and imbricated against each other at low angles on a
meter- to tens-of-meter scale. S0 is transposed into a
subvertical, N–S trending mylonitic foliation (S1 ) that
contains isoclinal intrafolial folds (F1 ) that refold S0
(Fig. 3a). Consequently, this composite transposition
fabric is referred to as S0 /S1 . S0 /S1 is an amphiboliteto retrograde greenschist-facies fabric that is defined by
metamorphic biotite, muscovite and chlorite in felsic
metavolcanics and clastic metasediments. Other metamorphic minerals such as kyanite, staurolite and hornblende, as well as plagioclase and quartz are deformed by
and exhibit a grain-shape preferred orientation parallel
to S0 /S1 . Associated with S0 /S1 is a pervasively developed, steep- to subvertical (60◦ –90◦ ) southerly plunging mineral stretching lineation (L1 ). L1 is defined by
rodded mineral aggregates and clasts, the grain-shape
preferred orientation of acicular mineral grains and by
aligned metamorphic mineral grains. The fold axes of
F1 folds are aligned and orientated parallel to L1 , resulting in a well-developed intersection lineation parallel
to L1 . Amphibolite- and retrograde amphibolite-facies
3.1. Fabric development in the TSB
The TSB is a 10-km long, 1-to 2-km wide N–S trending extremity of the BGB that occurs along the southern
margin of the belt (Figs. 1 and 2). It consists of variably deformed felsic metavolcanics and volcaniclastics,
amphibolite- and greenschist-facies metabasite, ultramafic rocks, chert and minor aluminous clastic metasediments of the Theespruit Formation (Viljoen and Viljoen,
1969a,b; Anhaeusser et al., 1981). The TSB is bounded
to the west by the 3445 ± 3 Ma Stolzburg Pluton (Kröner
et al., 1991) and to the east by the 3443 + 4/−3 Ma
Theespruit Pluton (Figs. 1 and 2; Kamo and Davis,
1994). Both the Stolzburg and Theespruit Plutons form
part of the early TTG suite that is intrusive into the
Theespruit Formation (Viljoen and Viljoen, 1969a; De
Wit et al., 1983).
3. Geology of the Tjakastad schist belt
the development of high-grade constrictional mylonitic
fabrics and post-peak metamorphic extensional shear
fabrics in the greenstones of the Stolzburg schist belt
(Kisters et al., 2003).
southern margin of the BGB, the transition between
the low-grade central portions of the belt and the highgrade gneisses and greenstone remnants is marked by
the Theespruit Formation, an allochtonous amphibolitefacies tectonic mélange of mafic and felsic volcanics
which occurs along the granitoid-greenstone contacts
(Viljoen and Viljoen, 1969b; De Wit et al., 1983). The
Theespruit Formation is separated from the rest of the
greenstone sequence by the Komati Fault (Viljoen and
Viljoen, 1969b; De Wit et al., 1983, 1987; Armstrong et
al., 1990; De Ronde and De Wit, 1994). The Komati Formation, a succession of lower greenschist-facies ultramafic lava flows, constitutes the stratigraphically lowest
part of the relatively intact greenstone sequence above
the Komati Fault (Viljoen and Viljoen, 1969b; Cloete,
1999; Dann, 2000).
Detailed structural and geochronological investigations have revealed that the BGB has experienced a
polyphase tectonic history and is made up of distinct
structural and stratigraphic domains that were assembled during two main accretionary episodes, D1 at ca.
3445 Ma and D2 at ca. 3230 Ma. (Williams and Furnell,
1979; Fripp et al., 1980; De Wit, 1982; De Wit et al.,
1983, 1987, 1992; Lowe et al., 1985, 1999; Lowe, 1994,
1999; De Ronde and De Wit, 1994; Kamo and Davis,
1994). Each of these episodes was accompanied by a
period of voluminous TTG magmatism (Fig. 1). D1 is
recognized in the southern parts of the BGB and affected
the lower formations of the Onverwacht Group (De Wit
et al., 1983, 1987). These formations were subjected
to an episode of subhorizontal thrusting and recumbent folding, during which the Komati Formation was
thrust onto the Theespruit Formation and synkinematically emplaced TTG basement along the Komati Fault
(De Wit et al., 1983, 1987; Armstrong et al., 1990; De
Ronde and De Wit, 1994). The main assembly of the
BGB occurred with the amalgamation of the southern
and northern blocks of the BGB during D2 at ca. 3230 Ma
(De Ronde and De Wit, 1994; Kamo and Davis, 1994;
De Ronde and Kamo, 2000).
However, it was recently recognized that peak metamorphism in the granitoid-gneiss terrain to the south of
the BGB occurred during D2 and post-dates the intrusion of the ca. 3.45 Ga TTG suite into this terrain by
more than 200 Ma (Dziggel et al., 2002). This implies
that the terrain as a whole was re-activated subsequent
to TTG emplacement and tectonically buried to midto lower crustal depths before exhumation and juxtaposition against the greenstone belt. Juxtaposition is
proposed to have been accomplished by extensional
detachment faulting related to the collapse of the D2 orogen, as evidenced by condensed metamorphic gradients,
91
strain regime. In the northern part of the study area,
the strongly prolate amphibolite-facies mylonites grade
into, and are overprinted by, greenschist-facies mylonites
away from the plutons and towards the central parts of
the BGB. Strain markers in these lower grade mylonites
Fig. 2. Geological map of the N–S trending Tjakastad schist belt that is bordered by the Stolzburg and Theespruit plutons. The position of sample
localities is also shown.
mylonites are characterized by strongly prolate (L S)
fabrics and L1 is the strongest fabric element developed in these rocks (Fig. 3b). Clasts and mineral aggregates have axial ratios of 20–100+:1–3:1, indicating that
fabric development occurred in a highly constrictional
92
A summary of the main petrographic characteristics
of the rocks of the TSB is presented in Table 1. The volumetrically dominant rock types of the Theespruit Formation consist of relatively simple mineral assemblages
that are not well suited to estimating metamorphic conditions. Consequently, the metamorphic investigation
focuses on the rare aluminous clastic metasediments and
garnet-bearing metabasic horizons that occur as intercalations within the more common rock types. These units
are discussed in more detail below and sample localities
are presented in Fig. 2.
3.3. Petrology and petrography
Pluton. This fabric is defined by rodded quartz and
quartz–plagioclase aggregates and plunges at steep to
subvertical angles (50◦ –80◦ ) to the east. Within ca. 50 m
of the granite-greenstone contact, the rodding fabric
grades to a subvertical gneissosity that is aligned parallel
to the granite-greenstone contact. The gneissosity is only
developed along the eastern margins of the Stolzburg and
Theespruit Plutons and is not as extensive or intense as
the gneissosity developed along the northern margin of
the Stolzburg Pluton (Kisters et al., 2003).
Fig. 3. Photographs illustrating the tectonization and fabric development of greenstones in the TSB. (a) Plan view (looking S) of a rootless isoclinal
fold (F1 ) that transposes bedding (S0 ) into the mylonitic S1 foliation in the central parts of the TSB. S1 trends N–S, parallel to the pen. (b) Oblique
view of an outcrop of felsic metavolcanics showing the pervasive subvertical rodding (L S) fabric typical of lithologies in the TSB.
have axial ratios that vary from 6:2:1 to 5:4:1, but are
generally close to plane strain. S-C fabric relationships
are extensively developed in these mylonites, indicating
non-coaxial shear during fabric development and that
shearing occurred in an extensional shear zone (e.g. Platt
and Vissers, 1980; Passchier and Trouw, 1996). S-C fabrics in the greenschist-facies mylonites and ␴-objects
and occasional S-C fabrics in the amphibolite-facies
mylonites indicate a (south)west-up–(north)east-down
sense of movement during deformation (Fig. 4a and b).
In the TSB, this corresponds to the upward displacement
of the Stolzburg pluton relative to the Theespruit pluton.
To the north of the Stolzburg pluton and northeast of the
Theespruit pluton S-C fabrics along the Komati schist
zone indicate that the TTG terrain has moved up relative
to the BGB.
3.2. Fabric development in the granitoids
Areas of the Stolzburg and Theespruit Plutons that
border on the TSB exhibit a persistent, although heterogeneously developed, steeply plunging mineral rodding
lineation similar to the fabric documented by Kisters et
al. (2003) along the northern margin of the Stolzburg
93
Fig. 4. S-C fabrics developed in the Theespruit Formation that consistently point to a (south)west side-up–(north)east side-down sense of displacement. (a) S-C fabric developed in chlorite schist within the Komati schist zone, ca. 500 m east of the Theespruit Pluton. Profile view looking SE
with the BGB on the left and the Theespruit pluton on the right. (b) S-C fabric developed in a felsic metavolcaniclastic unit ca. 100 m east of the
Stolzburg pluton in the TSB. Profile view looking S with the Theespruit pluton and BGB on the left and the Stolzburg pluton on the right.
Felsic volcaniclastics
Rock type
Dark green, coarse-grained, massive
to slightly schistose
Pale yellow, medium- to fine-grained,
schistose to mylonitic
Field occurrence
Peak: chl–ms–pl–qtz ± bt
Retrograde: chl Accessory: ilm
Mineral assemblages
Pl–qtz bands are dynamically recrystallized and intercalated with strongly foliated chl–ms bands
Coarse bands of aligned hbl are intercalated with thin, bands of recrystallized
pl–ep
Highly schistose, fabric defined by chl–ms
Texture
Table 1
Summary of the mineral assemblages and textural characteristics of the rocks of the TSB
Amphibolite-facies
metabasite
Peak: hbl–pl ± ep ± qtz
Retrograde: chl Accessory:
ttn–ilm
Chl–qtz ± ms ± zo Accessory:
ilm
Srp–cpx–chl–tlc–mgs–sd
Accessory: mag
Serpentinite consists of large cpx–srp
intergrowths; schists consist of alternating
tlc–chl foliae and mgs–sd bands
Coarse, poikiloblastic grt is enveloped by
chl–bt schistose fabric. St is aligned and
elongated in schistosity
Ky and st are aligned and elongated parallel to bt–ms foliation. Fibrous overgrowths of sil on ky
Poikilitic, sub- to euhedral grt in a matrix
of aligned hbl and fine- to mediumgrained pl–ep–qtz bands
Greenschist-facies
metabasite
Ultramafic rocks
Grt-bearing clastic
metasediment
Coarse-grained, massive to slightly
schistose
Peak: grt–st–bt–chl–pl–qtz
Retrograde: chl–ms Accessory:
ilm–ap–tur–aln
Peak: ky–st–bt–ms–pl–qtz
Retrograde: sil–chl Accessory:
tur–ilm–ap
Peak: grt–ep–hbl–pl–qtz
Retrograde: chl Accessory:
spn–ilm
Light green, fine-grained, schistose
to mylonitic
Pink, coarse-grained, massive
serpentinite bodies enveloped by
blue-green talc-chlorite schist
Medium-grained, highly schistose
intercalations associated with felsic
volcanics
Coarse-grained, foliated, associated
with felsic volcanics
Ky-bearing clastic
metasediment
Grt-bearing metabasite
Mineral abbreviations are after Kretz (1983).
94
Peak metamorphic minerals such as biotite, chlorite
and hornblende define the S1 /L1 fabric developed in the
rocks of the TSB. In addition, other peak metamorphic
porphyroblasts such as kyanite and staurolite are
deformed, elongated and aligned in S1 /L1 , while more
competent minerals (i.e., garnet) are enveloped by
S1 /L1 . The sum of this petrographic evidence suggests
that the main fabric-forming event in these rocks
occurred subsequent to the crystallization of the peak
metamorphic mineral assemblage and must, therefore,
post-date peak metamorphism. Staurolite, kyanite and
plagioclase crystals all behaved in a ductile manner
during post-peak metamorphic shearing, suggesting
that these minerals have undergone recrystallization
and, most likely, re-equilibration subsequent to peak
3.4. The relationship between mineral growth,
fabric development and mineral equilibration
tion of this sample involved a crossing of the kyanitesillimanite phase boundary.
Three samples of garnet-bearing metabasite were also
investigated as part of this study. The first two (Tj 3
and 62107) consist of 1–3 mm, subhedral, poikilitic garnet porphyroblasts contained in a matrix of mediumgrained, equigranular hornblende, epidote, plagioclase
and quartz. Quartz, ilmenite and epidote are present as
inclusions in garnet, while hornblende contains inclusions of ilmenite, titanite and quartz. Fine chlorite occasionally occurs as thin overgrowths replacing hornblende
and garnet. Hornblende is the main fabric-forming mineral in these rocks and the fabric is only slightly deflected
around the garnet porphyroblasts (Fig. 5g). The peak
metamorphic mineral assemblage in these samples is
considered to be grt–hbl–ep–pl–qtz. A third garnetbearing metabasite (sample 61406) differs from the
other two samples in that it contains large, euhedral
garnet porphyroblasts that have been pseudomorphed
by a pl–chl–hbl assemblage that forms distinct coronas
around the relic garnet core (Fig. 5h). The coronas also
contain small, anhedral garnet grains that are compositionally distinct from the main garnet porphyroblast. In
addition, the sample matrix consists of an assemblage of
fine-grained, aligned hornblende, plagioclase and quartz
that is extensively overprinted by large, randomly orientated chlorite grains. Sample 61406 comes from a
low-strain domain close to the Stolzburg pluton in the
northern part of the TSB (Fig. 2), which suggests that
it might have been shielded from the post-peak metamorphic deformation event that affected most of the
rocks in the TSB, thereby allowing the textures to be
preserved.
(1)
Clastic metasediments bearing the assemblage grt–
st–bt–chl–pl–qtz (mineral abbreviations follow Kretz,
1983) were sampled at three different localities in the
TSB. These samples (Tj 18, 62105F and 62601C) contain 2–5 mm rounded, subhedral, poikiloblastic garnet
grains (grt1) that contain abundant inclusions of quartz
and ilmenite. The foliation wraps around the garnet
porphyroblasts and all grains exhibit well-developed
pressure shadows. The pressure shadow sites predominantly contain quartz, but all three samples have grains
where the pressure shadows are filled by a generation of compositionally distinct garnet (grt2; Fig. 5a).
Biotite and chlorite (chl1) defines the S1 /L1 fabric in
these samples and staurolite porphyroblasts and plagioclase grains are deformed and aligned parallel to the
fabric, occasionally forming well-developed ␴-objects
(Fig. 5b–d). A second generation of compositionally
distinct chlorite (chl2) occurs as thin overgrowths on
certain biotite, chl1, grt1 and grt2 grains. The matrix
of these samples does not contain muscovite, and fine
muscovite is only present as a replacement product of
staurolite. Based on these observations it is proposed
that the peak metamorphic assemblage in these samples consists of grt–st–bt–chl–pl–qtz and formed via
the reaction
grt + chl + ms → st + bt + qtz + H2 O
(Holland and Powell, 1998), with muscovite as the limiting reactant in this reaction. The question of which of the
mineral generations in these samples represents the best
approximation of a chemically equilibrated assemblage
will be addressed in the following section. Sample Tj 18
contains a site where a garnet porphyroblast was broken
apart during deformation and the highly poikilitic garnet core was exposed to retrogression (Fig. 5e). This site
consists of fine intergrowths of garnet (grt3), plagioclase
(pl3), chlorite (chl2), muscovite and quartz that form an
assemblage potentially useful in constraining retrograde
metamorphic conditions.
A clastic metasediment containing the assemblage
ky–st–bt–ms–pl–qtz was sampled at one locality in the
TSB. Sample 62601D is from the same outcrop as sample 62601C and the two assemblages occur as alternating
bands within this metasedimentary horizon. The S1 /L1
fabric in this sample is defined by biotite and muscovite, with 1–3 mm kyanite and staurolite porphyroblasts present as elongated and micro-boudinaged grains
that are aligned in the fabric. Plagioclase is deformed
and occasionally forms ␴-objects, similar to plagioclase
in the other metasedimentary samples (Fig. 5d). Finegrained needles of sillimanite occur on certain kyanite
grains (Fig. 5f), suggesting that the metamorphic evolu-
95
Fig. 5. Photomicrographs. (a) Typical garnet porphyroblast from garnet-bearing metasediment displaying complex zoning consisting of a poikilitc
core—inclusion-free rim (grt1) and a second generation of garnet (grt2) that is confined to the pressure shadow sites (sample Tj 18, plane-polarized
light). (b) Thin section of garnet-bearing metasediment showing that the garnet porphyroblasts are enveloped by the bt–chl fabric in this sample,
while staurolite porphyroblasts are aligned and elongated parallel to the fabric (sample 62105F, plane-polarized light). (c) Staurolite-quartz ␴-object
illustrating the deformation and ductile recrystallization experienced by the peak metamorphic mineral assemblage in the metasediments (sample
62105F, plane-polarized light). (d) Plagioclase ␴-object indicating the ductile recrystallization of plagioclase in the metasedimentary samples (sample
62601D, crossed nicols). (e) Backscatter SEM image of a site of garnet breakdown in garnet-bearing metasediment sample Tj 18. The breakdown
assemblage consists of grt3, chl2, pl3, ms and quartz, and the mineral compositions presented in Table 2 were obtained from the area highlighted
by the box. (f) Overgrowths of fibrous sillimanite on kyanite grains in sample 62601D (plane-polarized light). (g) Typical texture of garnet-bearing
metabasite, consisting of a subhedral garnet porphyroblast in a matrix of hornblende, plagioclase, epidote and quartz. Aligned hornblende grains
define the fabric in these rocks and the fabric is slightly deflected around the garnet porphyroblasts (sample 62107, plane-polarized light). (h) Thin
section photograph of an euhedral garnet porphyroblast in garnet-bearing metabasite sample 61406, where the garnet has been pseudomorphed by a
corona of plagioclase-chlorite. The mineral compositions presented in Table 2 were obtained from the area highlighted by the box (plane-polarized
light). Mineral abbreviations are after Kretz (1983).
96
The mineral chemistry of the peak and retrograde
assemblages in the different samples from the TSB are
presented in Table 2. An investigation of the garnet-
4.1. Mineral chemistry
4. Conditions of metamorphism
have been in chemical equilibrium with only specific
portions of the garnet crystals, as discussed below.
Fig. 5. (Continued ).
metamorphism. Consequently, the current chemical
composition of these phases might not preserve peak
metamorphic compositions, but rather a composition
that re-equilibrated somewhere along the high-grade
portion of the retrograde path. Thus, this study proposes that the peak metamorphic assemblage persisted
during the higher grade portion of retrograde P–T path
evolution, but that, due to recrystallization, staurolite,
plagioclase and biotite were compositionally reset
during this event. These minerals are considered to
St
matrix
Grt3
g/br
Grt1
rim
Grt2
p/s
Tj 18 (grt-bearing clastic metasediment)
Grt1
core
29.63
0.62
56.15
0.35
0.28
1.13
11.55
0.42
0.10
98.47
3.98
8.89
1.30
0.02
0.11
0.04
0.02
0.04
0.06
14.46
Grt1 core
36.54
0.58
20.61
31.24
5.28
2.69
3.09
100.03
2.97
1.98
0.04
2.13
0.30
0.33
0.27
8.00
0.11
0.70
0.10
0.09
0.87
Grt1 rim
35.84
0.61
19.84
31.91
7.42
1.39
2.99
99.99
2.97
1.94
0.04
2.21
0.42
0.17
0.27
8.02
0.06
0.72
0.14
0.09
0.93
Grt2 p/s
SiO2
TiO2
Al2 O3
Cr2 O3
V 2 O5
ZnO
FeO
MnO
MgO
Total
Si
Al
Fe2+
Mg
Zn
Mn
V
Cr
Ti
Total
29.14
0.63
54.61
0.10
0.20
0.11
14.38
0.35
0.50
99.59
3.96
8.74
1.63
0.10
0.01
0.03
0.02
0.01
0.06
14.57
Bt
matrix
Chl1
matrix
39.64
1.36
18.17
0.12
18.28
0.14
12.45
0.00
10.26
100.43
2.83
1.17
0.36
0.07
0.01
1.09
0.01
1.32
0.93
7.80
54.83
Chl2
g/br
Pl2
matrix
60.22
25.16
6.94
7.57
0.11
100.00
2.68
1.32
0.33
0.65
0.01
4.99
0.34
0.66
Pl3 g/br
Chl1 matrix
30.30
0.09
26.01
0.14
26.80
0.23
16.32
0.00
0.12
100.00
2.76
1.24
1.55
0.01
0.01
2.04
0.01
2.21
0.00
0.01
9.84
52.04
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Fe2+
Mg
Na
K
Total
XPa
SiO2
Al2 O3
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
SiO2
58.84
Al2 O3 26.08
CaO
7.80
Na2 O
7.18
K2 O
0.10
Total 100.00
Si
2.63
Al
1.37
Ca
0.37
Na
0.62
K
0.01
Total
5.00
XAn
0.37
XAb
0.62
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
Na
K
Total
Mg#
30.14
0.02
23.19
0.00
31.93
0.45
12.44
0.00
1.30
100.09
2.85
1.15
1.43
0.00
0.00
2.57
0.03
1.75
0.00
0.16
9.94
40.50
Bt matrix
SiO2
30.03
TiO2
0.02
Al2 O3 25.30
Cr O
0.00
2
3
FeO
27.91
MnO
0.39
MgO
16.20
Na2 O
0.00
K2 O
0.63
Total 100.48
Si
2.75
Al IV
1.25
Al VI
1.48
Ti
0.00
Cr
0.00
Fe2+
2.14
Mn
0.02
Mg
2.21
Na
0.00
K
0.07
Total
9.92
Mg#
50.84
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
K
Total
Mg#
39.69
1.32
18.50
0.00
16.93
0.00
13.96
0.00
9.55
99.94
2.81
1.19
0.36
0.07
0.00
1.00
0.00
1.48
0.86
7.77
59.51
St matrix
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
K
Total
Mg#
62.71
23.41
4.59
9.21
0.08
100.00
2.78
1.22
0.22
0.79
0.00
5.01
0.21
0.78
97
Pl2 matrix
48.61
0.50
37.01
0.00
1.31
0.00
0.62
0.45
11.45
99.95
3.09
0.91
1.86
0.02
0.07
0.06
0.06
0.93
6.99
0.06
Ms g/ br
Table 2
Major element content and structural formulae of representative mineral analyses of the peak and retrograde assemblages in garnet-bearing metasediments and garnet-bearing metabasites
Sample
mineral
occurrence
37.21 37.44 36.62 SiO2
0.09
0.08
0.13 TiO2
20.94 20.94 20.71 Al2 O3
31.94 32.62 31.73 Cr2 O3
5.42
2.38
7.71 V2 O5
1.91
2.12
0.81 ZnO
2.48
4.42
2.32 FeO
100.00 100.00 100.01 MnO
3.03
3.02
3.02 MgO
2.01
1.99
2.01 Total
0.01
0.01
0.01 Si
2.17
2.20
2.19 Al
0.30
0.13
0.44 Fe2+
0.23
0.25
0.10 Mg
0.22
0.38
0.20 Zn
7.96
7.98
7.97 Mn
0.08
0.09
0.03 V
0.74
0.74
0.75 Cr
0.10
0.04
0.15 Ti
0.07
0.13
0.07 Total
0.90
0.90
0.96
36.08
1.08
19.81
32.17
5.07
2.25
3.54
100.00
2.96
1.91
0.07
2.21
0.29
0.27
0.31
8.02
0.09
0.72
0.09
0.10
0.89
62601C (grt-bearing clastic metasediment)
SiO2
36.92
TiO2
0.00
20.80
Al2 O3
FeO
31.43
MnO
7.13
MgO
1.67
CaO
2.05
Total
100.01
Si
3.02
Al
2.01
Ti
0.00
Fe2+
2.15
Mn
0.40
Mg
0.20
Ca
0.18
Total
7.97
XPy
0.07
XAlm
0.73
XSpss
0.14
XGrss
0.06
Fe/Fe + Mg
0.91
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Ti
Fe2+
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
98
Grt1 rim
37.81
0.24
20.24
30.17
6.86
1.60
3.08
100.00
3.08
1.94
0.01
2.05
0.39
0.19
0.27
7.94
0.07
0.71
0.13
0.09
0.91
Grt2 p/s
36.85
0.15
20.79
27.80
5.02
0.23
9.16
100.00
3.00
1.99
0.01
1.89
0.28
0.03
0.80
8.00
0.01
0.63
0.09
0.27
0.99
Grt rim
St matrix
29.30
0.55
55.19
0.29
0.21
0.20
13.57
0.53
0.18
99.31
3.97
8.81
1.54
0.04
0.02
0.05
0.02
0.03
0.06
14.53
41.87
0.71
14.76
0.00
23.24
0.40
4.44
11.40
2.13
1.03
99.98
6.33
1.67
0.96
0.00
0.08
1.00
2.94
0.00
0.00
0.02
1.85
0.14
0.49
0.20
15.69
25.38
Hbl matrix
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
K
Total
Mg#
Bt matrix
39.99
1.21
18.22
0.21
17.27
0.25
13.26
0.00
10.37
100.76
2.83
1.17
0.36
0.06
0.01
1.02
0.01
1.40
0.94
7.80
57.77
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
MnO
MgO
CaO
Total
Si
Al
Fe3+
Cr
Mn
Ca
Total
XPs
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
Na
K
Total
Mg#
40.28
0.19
26.95
0.03
8.66
0.20
0.00
23.69
100.01
3.07
2.42
0.50
0.00
0.01
1.94
7.94
0.17
Ep matrix
30.90
0.17
26.24
0.13
24.90
0.29
17.33
0.00
0.06
100.02
2.78
1.22
1.57
0.01
0.01
1.88
0.02
2.33
0.00
0.01
9.81
55.36
Chl1 matrix
SiO2
TiO2
Al2 O3
Cr2 O3
V2 O5
ZnO
FeO
MnO
MgO
Total
Si
Al
Fe2+
Mg
Zn
Mn
V
Cr
Ti
Total
Tj 3 (grt-bearing amphibolite)
36.80
0.00
20.80
30.60
5.81
2.22
3.97
100.20
2.99
1.99
0.00
2.08
0.33
0.27
0.35
8.01
0.09
0.69
0.11
0.11
0.89
62105F (grt-bearing clastic metasediment)
Table 2 (Continued )
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Ti
Fe2+
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Ti
Fe2+
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al IV
Al M123
Cr
Ti
Mg
Fe M123
Mn M123
Fe M4
Mn M4
Ca
Na M4
Na A
KA
Total
Mg#
SiO2
Al2 O3
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
Pl1
62.04
23.99
5.16
8.76
0.06
100.00
2.75
1.25
0.24
0.75
0.00
5.00
0.24
0.75
Pl2 matrix
60.83
24.73
6.18
8.14
0.12
100.00
2.70
1.30
0.29
0.70
0.01
5.00
0.29
0.70
Pl matrix
59.94
25.52
6.56
7.88
0.10
100.00
2.67
1.34
0.31
0.68
0.01
5.01
0.32
0.68
SiO2
Al2 O3
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
Table 2 (Continued )
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Ti
Fe2+
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Ti
Fe2+
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
Hbl matrix
44.08
0.43
16.74
0.06
18.68
0.30
6.65
11.75
0.99
0.35
100.02
6.42
1.58
1.29
0.01
0.05
1.44
2.21
0.00
0.06
0.03
1.83
0.08
0.20
0.07
15.27
38.79
Hbl corona
47.62
0.45
13.59
0.07
15.09
0.67
10.29
11.04
0.91
0.28
99.99
6.79
1.21
1.07
0.01
0.05
2.19
1.68
0.00
0.11
0.07
1.69
0.13
0.12
0.05
15.17
54.85
SiO2
Al2 O3
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
MnO
MgO
CaO
Total
Si
Al
Fe3+
Cr
Mn
Ca
Total
XPs
62.71
23.49
4.59
9.14
0.08
100.00
2.78
1.23
0.22
0.78
0.00
5.01
0.22
0.78
Pl corona
40.09
0.20
28.15
0.14
7.16
0.47
0.00
23.80
100.00
3.05
2.52
0.41
0.01
0.02
1.94
7.96
0.14
Ep matrix
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al IV
Al M123
Cr
Ti
Mg
Fe M123
Mn M123
Fe M4
Mn M4
Ca
Na M4
Na A
KA
Total
Mg#
62107 (grt-bearing amphibolite)
Grt rim
37.13
0.08
20.97
27.04
4.99
0.97
8.85
100.02
3.00
2.00
0.01
1.83
0.28
0.12
0.77
7.99
0.04
0.61
0.09
0.26
0.94
Grt corona
61406 (grt-bearing amphibolite)
36.74
0.14
21.19
25.33
9.92
1.31
5.37
100.00
3.00
2.04
0.01
1.73
0.56
0.16
0.47
7.97
0.05
0.59
0.19
0.16
0.92
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al IV
Al M123
Cr
Ti
Mg
Fe M123
Mn M123
Fe M4
Mn M4
Ca
Na M4
Na A
KA
Total
Mg#
SiO2
Al2 O3
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Cr
Fe2+
Mn
Mg
Na
K
Total
Mg#
99
Pl matrix
58.14
26.65
8.16
6.98
0.08
100.00
2.60
1.40
0.39
0.60
0.00
5.00
0.39
0.61
Chl corona
31.92
0.13
24.87
0.14
22.14
0.39
20.37
0.00
0.04
100.01
2.84
1.16
1.45
0.01
0.01
1.65
0.02
2.70
0.00
0.00
9.84
62.12
Structural formulae were calculated on the basis of 12 oxygens for grt, 23 for st and hbl, 11 for bt and ms, 14 for chl, 8 for pl and 13 for ep.
XPy, XAlm, XSpss and XGrss as defined by Spear (1993). XAn = Ca/(Ca + Na), XAb = Na/(Ca + Na), XPa = Na/(Na + K), XPs = Fe3+ /(Fe3+ + Al),
Mg# = 100 × Mg/(Mg + Fe) p/s = pressure shadow of grt1 porphyroblast; g/br = garnet breakdown texture in Fig. 5e; corona = garnet breakdown
texture in Fig. 5h.
100
either within single crystals or between different crystals
in the same sample. All samples contain two generations
of chlorite, with chl1 having significantly higher XMg
than chl2 (Table 2). Plagioclase does not show compositional variations within single grains or between grains
in the same sample, but compositional differences, likely
related to bulk composition, do exist between the samples. A notable exception to the above is sample 62105F,
as it contains two distinct generations of plagioclase. The
Fig. 6. Major and transition trace element plots along a traverse through a garnet from garnet-bearing metasedimentary sample Tj 18. The zonation
exhibited by this traverse is typical of garnet in these samples and suggests two distinct episodes of garnet growth. The first growth episode (grt1,
unshaded areas) likely occurred under prograde conditions, while the second growth episode (grt2, shaded areas) is confined to the pressure shadow
sites of the grt1 grains. Grt2 is characterized by a two- to threefold increase in Cr and V concentrations and higher XCa content relative to grt1. The
line of the traverse is indicated on the photograph and sketch and all traverses are plotted from A to B.
bearing metasediments revealed that, in all three samples, grt1 porphyroblasts exhibit a core-to-rim chemical
variation with typical prograde zonation patterns of Mn,
Fe and Mg (Fig. 6; e.g. Spear, 1993). The composition
of garnet that grew in the pressure shadows of the garnet
porphyroblasts (grt2), exhibits a marked increase in Ca
as well as pronounced Cr–V enrichment relative to grt1
(Fig. 6). Staurolite and biotite have very uniform compositions and no compositional variations were observed
101
P–T conditions were estimated using the program
THERMOCALC (Version 3.21; Holland and Powell,
1998) and the internally consistent dataset of Holland
and Powell (1998; incorporating subsequent upgrades).
All calculations were performed in ‘Average P–T’
mode and assumed the presence of a pure water fluid
phase. Mineral end-member activities were calculated
at 7.5 kbar and 550 ◦ C with the program AX (Holland
and Powell, 1998). P–T estimates from garnet-bearing
4.2. Estimation of peak metamorphic conditions
contact, or minerals that appear to have been in mutual
contact during retrogression. The approach of the rocks
to equilibrium was evaluated by considering the consistency of Fe/Mg KD values for the ferromagnesian
mineral assemblage in rocks of different composition
from the same outcrop area. The consistent KD values
from these assemblages were taken to suggest a reasonable approach to equilibrium.
Garnet-bearing metabasite samples targeted for thermobarometric calculations contain a peak metamorphic
assemblage that consists of garnet and single generations
of hornblende, plagioclase, epidote and quartz. Garnet
is not strongly zoned and does not display evidence of
multiple garnet growth events, suggesting that all garnet growth occurred during a single prograde episode
within a restricted P–T range. Retrograde garnet growth
did not occur in the metabasites, as these rocks did not
experience a retrograde garnet-producing reaction such
as the reaction that involves plagioclase and staurolite
re-equilibration in metasediments. Plagioclase is recrystallized, although not to the same extent as in metasedimentary samples and multiple generations of plagioclase
are not present in these rocks. The single generations of
matrix minerals in samples Tj 3 and 62107 were paired
with garnet rim compositions as a best approximation of
a peak metamorphic equilibrium assemblage. Thermobarometric calculations performed on sample 61406 are
aimed at estimating retrograde metamorphic conditions
and, therefore, the minerals that formed at the site of garnet breakdown in this sample (Fig. 5h) were paired as an
approximation of an equilibrated retrograde assemblage.
bulk of plagioclase in this sample has been recrystallized during deformation and has a composition of ca.
An24 (pl2). However, certain low-strain lithons within
the matrix of this sample contain plagioclase with a composition of ca. An32 (pl1). This An-rich plagioclase is
interpreted to preserve a peak metamorphic composition as it escaped recrystallization and re-equilibration
during post-peak metamorphic shearing. The pervasive
re-equilibration of plagioclase from peak metamorphic
pl1 compositions to pl2 compositions in sample 62105F
suggests that all the deformed minerals in the garnetbearing metasediments likely experienced significant reequilibration subsequent to peak metamorphism. Therefore, it is most probable that the single, uniform composition exhibited by the recrystallized peak assemblage
minerals in these rocks does not reflect their peak metamorphic composition, but rather a composition attained
during the higher grade portion of the retrograde P–T
evolution. The Ca that was released from plagioclase during re-equilibration is likely to have been incorporated
into another Ca-bearing mineral, such as garnet, that was
crystallizing at that time. The petrographic context of
grt2 suggests that its crystallization was syn-tectonic,
and the pronounced Ca enrichment of grt2 suggests
that it might have crystallized during plagioclase reequilibration. Similarly, the Cr–V enrichment in grt2 is
most likely caused by the re-equilibration or breakdown
of a Cr–V-bearing phase (such as staurolite) under these
conditions. The reaction whereby the peak metamorphic
compositions were re-equilibrated probably involved a
partial reversal of reaction (1), resulting in the consumption of peak composition staurolite, biotite and pl1 to
produce the high-grade retrograde compositions of grt2,
pl2, staurolite and biotite. Consequently, the best approximation of the near-peak-metamorphic, chemically equilibrated assemblage in the garnet-bearing metasediments
consists of grt2 paired with staurolite, biotite, pl2, chl1
and quartz (Table 3). As a result, peak and prograde
metamorphic conditions, although reflected by the chemistry of portions of the garnet crystals, are unresolvable
by geothermobarometric techniques that rely on mineral chemistry. The mineral compositions paired in the
P–T calculations are in all cases from minerals in mutual
Sample
Grt1 (core)
Grt1 (core)
Prograde P–T conditions
Grt1 (rim)
Grt1 (rim)
Grt1 (rim), pl1
Peak P–T conditions
Grt2, st, bt, pl2, chl1
High-grade retrograde P–T conditions
Grt3, chl2, pl3, ms
Late retrograde P–T conditions
Table 3
Summary of the different mineral generations present in the garnet-bearing metasedimentary samples and the likely point along the P–T evolutionary
path where each generation attained its measured chemical composition (as presented in Table 2)
Tj 18
62601C
62105F
102
P (kbar)
1.1
0.9
1.0
1.6
1.2
±
569
543
556
563
569
556
537
T (◦ C)
42
20
19
14
18
59
45
±
0.186
0.596
0.152
0.125
0.290
0.952
0.958
Corr
2.81
1.53
1.35
0.90
1.16
1.57
0.78
Fit
7
5
7
7
7
7
6
Na
Assemblage
Sample
7.9
7.7
7.2
7.0
7.0
2.7
1.3
Table 4
THERMOCALC results of peak and retrograde P–T conditions calculated from the equilibrated assemblages in garnet-bearing metasedimentary
and garnet-bearing metabasite samples
Peak metamorphic conditions
62105F
Grt2–st–bi–chl1–pl2–qtz
62601C
Tj 18
62107
Grt–hbl–pl–ep–qtz
Tj 3
Grt–hbl–pl–ep–qtz
6.1
3.8
Fig. 8. P–T diagram showing the position of the aluminosilicate-in
reaction in low-Al metapelites (left) and the staurolite-out reaction
(right) and the stability fields of ky–st and sil–st coexistence between
these two reactions (after Holland and Powell, 1998). Path A indicates the transition from ky–st coexistence to sil–st coexistence in
response to increasing temperature below ca. 7 kbar. Path B illustrates
that this transition cannot occur above ca. 7 kbar in response to increasing temperature, but must involve some component of decompression,
as illustrated by path C. Minimum pressure estimates of ca. 7.4 kbar
from the TSB indicate that the ky–st to sil–st transition observed in
sample 62601D must have followed a decompression path similar
to C.
temperatures in the TSB by ca. 20–40 ◦ C. The minimum
pressure estimates obtained from the TSB are supported
by experiments that constrain the garnet-in reaction in
compositionally similar metabasite to occur between
ca. 8 and 10 kbar at 550 ◦ C (Poli, 1993).
N is the number of independent equilibria calculated for each assemblage.
Retrograde metamorphic conditions
61406
Grt–chl–hbl–pl–qtz
Tj 18
Grt3–chl2–ms–pl3–qtz
a
metasediments and garnet-bearing metabasites constrain
P–T conditions of 7.4 ± 1.0 kbar and 560 ± 20 ◦ C in
the central parts of the TSB (Table 4 and Fig. 7). These
estimates reflect P–T conditions during high-grade
retrogression and are in all likelihood a conservative
minimum estimate of peak metamorphic conditions.
Sample 62601D contains co-existing kyanite and staurolite, which occurs over a fairly restricted temperature
window at 580–640 ◦ C (Fig. 8). This suggests that the
temperature estimates obtained by thermobarometry
could underestimate the actual peak metamorphic
Fig. 7. P–T plot of THERMOCALC estimates of peak and retrograde
metamorphic conditions. Point A represents peak metamorphic estimates from garnet-bearing metabasites (circles) and garnet-bearing
metasediments (squares). The retrograde path is constrained by point
B, the P–T estimate from the garnet breakdown assemblage in sample
61406 (Fig. 5h); point C, the P–T estimate from the garnet breakdown assemblage in sample Tj 18 (Fig. 5e) and point D, inferred from
the replacement of biotite by chlorite-muscovite assemblages in all
metasedimentary samples. Error ellipses are at two standard deviations
and the aluminosilicate phase diagram is after Holdaway (1971).
103
Fig. 10. U–Pb concordia diagram for titanite from sample Tj 3. Error
ellipses are at one standard deviation and the upper intercept age calculation is reported with 2σ error.
would place a maximum constraint on the timing of both
metamorphism and deformation in these rocks. Titanite
appears to be homogenous in major element composition
and no evidence of zonation could be found by optical
examination or backscatter electron imaging. This was
confirmed by time-resolved acquisition signals obtained
during LA–ICP-MS analysis that displayed fairly constant isotopic ratios during ablation and data acquisition.
The titanite is low in U, and consequently, fairly large
errors are associated with the isotopic measurements
(Table 5) that propagate through to relatively large uncertainty in the age determination (Fig. 10). Titanite from
sample Tj 3 yields an upper intercept concordia age of
3229 ± 25 Ma that is interpreted as the crystallization
age of titanite in this sample. The closure temperature
of titanite (ca. 650 ◦ C for grains as small as 200 ␮m;
Cherniak, 1993; Scott and St. Onge, 1995; Verts et al.,
1996; Frost et al., 2000) is grain-size dependent but, considering the ca. 560 ◦ C temperature estimates from the
TSB, it is unlikely that peak metamorphic temperatures
exceeded the closure temperature of these grains. The
Fig. 9. Photomicrograph of titanite grains in sample Tj 3, illustrating
the paragenesis of titanite dated in this sample. Grains occur as inclusions in peak metamorphic hornblende and are aligned parallel to the
S0 /S1 fabric. Plane-polarized light.
4.3. Retrograde metamorphic conditions
Petrologic indications of decompression during
retrogression are provided by the occurrence of
plagioclase-chlorite coronas that pseudomorph after
garnet in metabasite (Fig. 5h; e.g. Hoschek, 2001) and by
fibrous sillimanite overgrowths on kyanite in metasediments (sample 62601D; Fig. 5f). Staurolite in sample
62601D does not show any evidence of destabilization
and appears to coexist with both kyanite and sillimanite.
The P–T fields of kyanite-staurolite and sillimanitestaurolite coexistence are shown in Fig. 8. This figure
illustrates that the transition from kyanite-staurolite to
sillimanite-staurolite stability most likely coincided with
decompression, as pressure estimates from the TSB are
too high for this transition to have occurred simply by an
increase in temperature. The mineral assemblages that
formed at sites of garnet breakdown in samples 61406
(Fig. 5h) and Tj 18 (Fig. 5e) were used to obtain estimates
of retrograde P–T conditions (Table 4 and Fig. 7). These
estimates suggest that the rocks of the TSB experienced
near-isothermal decompression of ca. 4 kbar prior to
cooling. The replacement of biotite by chl2-muscovite
assemblages occurs in most of the metasedimentary
samples investigated and indicates that cooling and
retrogression continued into the greenschist-facies to
below the biotite isograd (Fig. 7; Ferry, 1984).
4.4. Timing of peak metamorphism
Different models for the evolution of the BGB have
proposed that peak metamorphism in the Theespruit Formation occurred during D1 at ca. 3.45 Ga (De Wit et al.,
1983; Armstrong et al., 1990; De Ronde and De Wit,
1994) or during D2 , contemporaneous with peak metamorphism in the spatially associated granitoid-gneiss
terrain (Kamo and Davis, 1994; Dziggel et al., 2002;
Kisters et al., 2003). This suggests that the timing of
peak metamorphism in the TSB is likely to coincide with
either D1 or D2 . An attempt was made to constrain the
timing of peak metamorphism in the TSB by in situ dating of titanite from garnet-bearing metabasite sample Tj
3. Titanite in this sample is present as 30–70 ␮m long,
diamond-shaped, yellow-brown translucent grains that
are present as inclusions in hornblende and in the matrix
of the sample (Fig. 9). The diamond-shaped titanites are
parallel to the grain-shape preferred alignment of hornblende and recrystallized quartz-plagioclase aggregates
of the matrix that define the main fabric in the rock.
As titanite is present as inclusions in peak metamorphic
hornblende and is aligned parallel to the S0 /S1 fabric
in this sample, the age obtained from this generation
104
±
0.7687
0.5712
0.5317
0.7352
0.7878
0.6582
0.7666
0.5734
0.4706
0.5794
0.6658
0.7197
206 Pb/238 U
0.0512
0.0690
0.0565
0.0259
0.0459
0.0861
0.0300
0.0377
0.0558
0.0491
0.0984
0.0332
±
0.2542
0.2673
0.2706
0.2416
0.2443
0.2647
0.2419
0.2664
0.3018
0.2726
0.2541
0.2362
207 Pb/206 Pb
0.0191
0.0353
0.0237
0.0073
0.0254
0.0171
0.0093
0.0188
0.0240
0.0148
0.0268
0.0160
±
3381
3141
3084
3288
3367
3269
3330
3142
3071
3174
3240
3245
207 Pb/235 U
3676
2913
2749
3553
3745
3260
3668
2922
2486
2946
3290
3495
206 Pb/238 U
3211
3290
3310
3130
3148
3275
3133
3285
3479
3321
3210
3094
207 Pb/206 Pb
Age calculations (Ma)
207 Pb/235 U
Isotopic ratios
1.471
2.036
1.129
0.751
1.346
1.689
0.898
1.042
1.059
1.098
2.668
1.399
Table 5
U–Pb isotopic data and calculated ages obtained from titanite in sample Tj 3
Spot
26.945
21.052
19.841
24.491
26.536
24.019
25.570
21.063
19.580
21.781
23.322
23.437
Pressure estimates of 8–11 kbar determined in
supracrustal remnants from the southern granitoidgneiss terrain (Dziggel et al., 2002) indicate a burial of
these rocks to mid- and lower crustal depths of 30–35 km.
This assumes an overlying mafic–ultramafic crust with
an average density of ρ = 3.0 g/cm3 (Cloete, 1999), but
since large parts of the crustal profile were probably sialic
in composition, a burial of the granitoid-gneiss terrain to
depths of ≥35 km seems likely. The minimum pressure
estimates of ca. 7.5 kbar from the TSB are slightly lower
than this and suggest the burial of the Theespruit Formation to depths of at least ca. 25 km. Significantly, the highP, low-T metamorphism in the Theespruit Formation
5.1. Implications for Mesoarchaean geothermal
gradients
structural manifestation of presumably higher heat flows
and more pronounced density contrasts in the Archaean,
in that both the increased ductility of rocks and the
enhanced density contrasts between the plutonic and
supracrustal rocks suites may have a led to a buoyancycontrolled overturn of the crust (e.g. Collins et al., 1998;
Marshak, 1999; Chardon et al., 2002; Van Kradendonk,
2004).
The lithologies and mineral assemblages studied from
the Theespruit Formation allow, for the first time in this
terrain, detailed P–T and geochronological constraints
on the tectono-metamorphic evolution of the granitoidgreenstone contacts in the southern parts of the BGB.
These data shed light on the apparent geothermal gradients in this Mesoarchaean crustal segment and the
implications these may have for the structural processes
that led to the amalgamation of the Barberton granitoidgreenstone terrain.
sph 1
sph 2
sph 3
sph 4
sph 5
sph 6
sph 7
sph 8
sph 9
sph 10
sph 11
sph 12
metamorphic age obtained from the TSB correlates very
well with other, more precise ages that constrain the timing of peak metamorphism in the southern Barberton
terrain at ca. 3230 Ma (Kamo and Davis, 1994; Dziggel
et al., 2002; Kisters et al., 2003). In other words, peak
metamorphism in the TSB occurred during D2 and postdates D1 and the intrusion of the southern TTG suite by
ca. 220 Ma.
5. Discussion
Several tectonic studies on various Archaean cratons
have identified distinct structural differences between
Archaean granitoid-greenstone terrains and younger
orogenic belts of, e.g., Proterozoic or Phanerozoic
provinces that have been suggested to represent secular
changes in orogenic style throughout the Earth’s history
(Anhaeusser, 1984; Choukroune et al., 1995; Hamilton,
1998; Van Kradendonk, 2004). Most of these structural differences are attributed to presumably higher heat
flows in the Archaean which resulted in the rheological
weakening of crustal rocks. As a consequence, significant crustal stacking and overthickening, as is commonly
observed in more recent collisional belts, are unlikely to
have been achieved in Archaean orogens because the
weakened crustal column could not have sustained large
vertical loads (e.g. Marshak, 1999; Chardon et al., 2002).
This may not only explain the lack of high-pressure rocks
from the Archaean rock record, but also the pronounced
vertical component of tectonic structures and kinematics in the evolution of Archaean granite-greenstone terrains. Indeed, numerous models suggest vertically driven
tectonics to have dominated the structural evolution of
Archaean cratons. The typical dome-and-keel structures
of Archaean terrains are sometimes interpreted to be the
105
Formation indicate that this crustal section was already
assembled at ca. 3450 Ma (e.g. Kamo and Davis, 1994).
In other words, the southern granitoid-gneiss terrain and
the Theespruit Formation behaved as a coherent entity
during the ca. 3230 Ma mid- to deep-crustal burial of the
rocks.
Sections of the Pilbara Craton in western Australia and the BGB show numerous regional-scale similarities, such as the localized high-pressure, low- to
Fig. 11. P–T diagram and map of the southern part of the Barberton granitoid-greenstone terrain summarizing the available metamorphic P–T data
for this area. Peak P–T estimates from the TSB (point A), the high-grade gneiss terrain (point B; Dziggel et al., 2002) and the Stolzburg schist belt
(point C; Kisters et al., 2003) were attained during D2 at ca. 3.23 Ga and record very similar apparent geothermal gradients of ca. 20 ◦ C/km. Point
D represents likely P–T conditions in the Komati Formation during D2 (Cloete, 1999) and point E indicates peak metamorphic conditions in the
Komati Formation that were likely attained during D1 (Lopez-Martinez et al., 1992; Cloete, 1999). Note the contrasting styles of metamorphism
in the Theespruit Formation and gneiss terrain compared to the Komati Formation as well as the position of the D2 extensional detachment that
separates these two terrains. ST = Stolzburg Pluton, TP = Theespruit Pluton.
and the southern granitoid-gneiss terrain records very
similar apparent geothermal gradients of ca. 20 ◦ C/km
(Fig. 11). Peak metamorphic conditions in both areas
were attained at ca. 3230 Ma, corresponding to the main
phase of the D2 deformation in the belt and post-dating
the intrusion of the TTG suite by ca. 220 Ma (De Ronde
and De Wit, 1994; De Ronde and Kamo, 2000). However, the clearly intrusive contact relationships between
the Stolzburg and Theespruit plutons and the Theespruit
106
The metamorphic conditions of the southern
granitoid-gneiss terrain contrast dramatically with the
low-P, moderate-T greenschist-facies metamorphism
exhibited by the Komati Formation in the lower parts
of the BGB (Fig. 11; Cloete, 1999). Peak metamorphic
conditions in the Komati Formation were most likely
attained during D1 at ca. 3450 Ma (Lopez-Martinez et
al., 1992), but the supracrustal sequence only underwent regional, lower greenschist-facies metamorphism
during D2 at ca. 3230 Ma for which van Vuuren and
Cloete (1995) and Cloete (1999) have estimated peak
pressures of ca. 2.6 ± 0.5 kbar. D2 pressure estimates
from the Theespruit and Komati Formations indicate that
a metamorphic break equivalent to a crustal column of
ca. 18 km separated these formations at that time. Therefore, even though the Theespruit and Komati Formations
might have experienced a shared TTG magmatic history
at 3.45 Ga (De Wit et al., 1987; Armstrong et al., 1990;
Kamo and Davis, 1994), the contrasting metamorphic
evolution of the two terrains during D2 indicates that
their final juxtaposition could not have occurred prior to
ca. 3230 Ma.
Both the strain intensity, style of deformation and
the metamorphic conditions described here for the TSB
highlight the significance of the Theespruit Formation
5.2. Tectonic rates
mal boundary layer at or prior to 3.2 Ga in this part of
the Kaapvaal Craton. This early stage of lithospheric
coupling between the crust and a mantle keel to form
stabilized cratonic nuclei is also indicated by the oldest
formation of diamonds in the central–western parts of
the Kaapvaal Craton (Shirey et al., 2003).
The fact that the continental crustal section exposed in
the southern granitoid-gneiss terrain was evidently sufficiently strong to undergo rapid burial to lower-crustal
depths of >35 km is in agreement with the low apparent geothermal gradients. The low-strain intensity and
structural integrity exhibited by, e.g., the Stolzburg and
Theespruit plutons illustrate the rigidity of this crustal
segment that has escaped pervasive ductile flow during burial as well as exhumation. These results indicate
that the Mesoarchaean crust in the southern Barberton
granitoid-gneiss terrain was able to sustain substantial
tectonic thickening, pointing to relatively cool and previously stabilized crust. Moreover, the preservation of
such low apparent geothermal gradients suggests that
crustal thickening and subsequent exhumation occurred
rapidly and that the rocks of the high-grade terrain did
not reach thermal equilibrium. The possible rates of tectonism are discussed below.
medium-temperature metamorphism of the supracrustal
sequences and the typical dome-and-keel geometries
between TTG domes and greenstone synforms (Collins
et al., 1998; Van Kranendonk et al., 2002; Van Kradendonk, 2004). Tectonic models proposed to account for
the high-grade metamorphism of greenstone sequences
in the Pilbara Craton critically hinge on the presumably high rate of heat production in the Mesoarchaean
crust that is assumed more than double compared to
modern day rates (e.g. Sandiford and McLaren, 2002).
This, in turn, has probably led to a significant weakening of crustal rheologies (e.g. Marshak, 1999). The
high-grade metamorphism, special thermal regime and
widespread constrictional-type strains recorded in these
supracrustals are interpreted to indicate the gravitational
sinking and burial of denser, mainly mafic and ultramafic
greenstone rocks along the flanks of, and between, buoyant and rising solid-state TTG diapirs, a model known as
partial convective overturn of the crust (e.g. Collins et al.,
1998; Van Kranendonk et al., 2002). However, in rocks
of the TSB and along the southern granitoid-greenstone
contacts of the BGB, the high-grade and mainly prolate
fabrics are exhumation fabrics. They are not related to the
burial or sinking of the supracrustal sequence. Moreover,
the highest pressures (8–11 kbar; Dziggel et al., 2002)
are documented from the southern TTG terrain and not
in the greenstone sequences. Since the felsic plutonic
rocks contain metamorphic assemblages recording significantly higher pressures than the flanking greenstones,
the sinking model of the greenstones between buoyantly
rising TTG diapirs appears unlikely. We envisage crustal
stacking and thickening to have occurred in a collisional
setting, concurrent with the main phase of D2 terrain
accretion recorded in the central parts of the shallowcrustal greenstone belt. This interpretation is consistent
with the crustal stacking depicted in deep reflection seismic profiles across the central Kaapvaal Craton (De Wit
and Tinker, 2004).
The remarkably low apparent geothermal gradients of
ca. 20 ◦ C/km preserved in both the southern granitoidgneiss terrain and the Theespruit Formation have, to our
knowledge, not been documented from a Mesoarchaean
high-grade metamorphic terrain and are clearly at variance with models of high Archaean heat flow and heat
production. These values are, in contrast, comparable
with the apparent geothermal gradients reported from
modern continental orogenic environments and subduction zones (e.g. Spear, 1993). The reason for the low
apparent geothermal gradients documented in this study
may be sought in the low amount of heat-producing
radiogenic elements (K, U, Th) in the 3.45 Ga TTG suite
(Yearron, 2003). It also suggests the presence of a ther-
107
108
Appendix A. Analytical techniques
2+ “S” instrument coupled to an in-house built 266 nm
NdYAG laser to measure Pb/U, Th/U and Pb isotopic
ratios. The sample introduction system was modified
to enable simultaneous nebulisation of a Tl–Bi–U–Np
tracer solution and laser ablation of the solid sample.
Natural Tl (205 Tl/203 Tl = 2.3871, Dunstan et al., 1980),
209 Bi and enriched 233 U and 237 Np (>99%) were used
in the tracer solution, which was aspirated to the plasma
in an argon–helium carrier gas mixture through an Glass
Expansion Micromist nebuliser, Scott-type double-pass
spray chamber and a T-piece tube attached to the back
end of the plasma torch. A helium gas line carrying the
sample from the laser cell to the plasma was also attached
to the T-piece tube.
The laser was set up to produce energy of 0.1–
0.5 mJ/pulse (measured just before the beam entered
the objective of the microscope) at a repetition rate of
10 Hz. The laser beam was focused ca 100–200 ␮m
above the surface of the sample in a 16.5 cm3 cell, which
was mounted on a computer-driven motorized stage of
a microscope. During ablation, the stage was moved
beneath the stationary laser beam to produce a square
laser pit in the sample. The size of the raster varied
between 20 ␮m × 20 ␮m and 40 ␮m × 40 ␮m to match
available titanite grain size and the depth of pits between
5 and 30 ␮m. Typical acquisitions consisted of a 60 s
measurement of analytes in the gas blank and aspirated
solution, followed by measurement of U, Th and Pb signals from accessory minerals, along with the continuous
203 Tl, 205 Tl, 209 Bi, 233 U and 237 Np signal from the aspirated solution, for another 180 s. The data were acquired
in time-resolved – peak jumping – pulse counting mode
with 1 point measured per peak for masses 202 (Hg), 203
(Tl), 205 (Tl), 206 (Pb), 207 (Pb), 208 (Pb), 209 (Bi), 232
(Th), 233 (U), 238 (U) and 237 (Np). Oxides at masses
248, 249, 253 and 254 were monitored. Quadrupole settling time was 1 ms for all masses except masses 231 and
233 where the settling time was increased to 10.3 ms by
inserting a dummy mass before the measured peaks at
masses 231 and 233 in each sweep to improve precision
of 232 Th. The dwell time was 8.3 ms on each mass except
for mass 207 where it was 24.9 ms. Raw data were corrected for dead time of the electron multiplier (20 ns)
and processed off-line in a spreadsheet-based program
(LAMDATE, Kosler et al., 2002; ISOPLOT, Ludwig,
1999). Data reduction included correction for gas blank,
laser-induced elemental fractionation of Pb, U and Th
and instrument mass bias.
To monitor accuracy and precision during data acquisition, the LAC titanite was analyzed (Pedersen et al.,
1988). This in-house standard is a several centimetres
large grain from a pegmatite in the Lillebukt Alkaline
stoichiometries and the resultant mineral structural formulae were used to evaluate the quality of the analytical
data. Only analyses that produced cation totals within
0.05 to 0.1 (depending on the number of cations) of
the ideal stoichiometric number for each cation site
were considered. For example, garnet: Si = 3 ± 0.05;
Al = 2 ± 0.05; Fe + Mg + Mn + Ca = 3 ± 0.1 and plagioclase: Al + Si = 4 ± 0.03 and Ca + Na = 1 ± 0.03. Mineral
standards with a known chemical composition were
treated as unknowns and analyzed using the same instrument and calibration set-up employed for unknowns.
For example, a plagioclase standard not used in the
plagioclase analytical routine was analyzed and the
measured chemical composition was then compared to
the actual, published composition. A comparison of the
measured and actual chemical compositions of selected
mineral standards used to test the accuracy of the analytical technique is presented in Table A.1 and serves as a
reflection of the absolute error associated with the major
element mineral compositions determined during this
study. The mineral compositions presented in this paper
represent the average of at least three repeat analyses of
the same mineral generation in a single sample.
A.2. Trace element mineral chemistry
Transition trace element concentrations in garnet
grains from a garnet-bearing metasedimentary sample
(Tj 18) were determined by laser ablation-inductively
coupled plasma–mass spectrometry (LA–ICP-MS) at
the Memorial University of Newfoundland. A detailed
description of this instrument, the analytical routine and
data processing techniques used in the trace element
analysis is presented in Horn et al. (1997). Analyses
were calibrated against USGS reference material BCR2G and Si (measured by electron microprobe) was used
as the internal standard. BCR2-G was also routinely
analyzed during the procedure to monitor instrument
stability.
This material is based on work supported by the NRF
under grant numbers NRF 2053186 and NRF 2060045.
Land-owners in the Tjakastad area are thanked for access
to their land and for their hospitality during field work.
Constructive reviews by Martin van Kranendonk, Dirk
van Reenen and an anonymous reviewer are greatly
appreciated.
A.1. Major element mineral chemistry
In situ U–Pb analysis of titanite was performed by
LA–ICP-MS (Jackson et al., 1996; Horn et al., 2000;
Kosler et al., 2002; Kosler and Sylvester, 2003; Tiepolo,
2003; Rawlings-Hinchey et al., 2003; Fonneland et al.,
2004; Hodych et al., 2004) at the Memorial University of Newfoundland and was carried out on 100-␮m
thick petrographic thin sections. Isotopic analysis of the
titanites by laser ablation ICP-MS followed a modified
technique described in Kosler et al. (2002) and Cox et al.
(2003). Analyses were performed on a VG PlasmaQuad
A.3. U–Pb geochronology
Major element mineral chemistry analyses were performed on a LEO 140VP scanning electron microscope
coupled to a Link ISIS energy dispersive spectrometry
system at the University of Stellenbosch. The data
presented in this paper are among the first to be
generated by this facility and, therefore, details of the
analytical procedure are included. The microscope was
operated at 20 kV with a beam current of 120 ␮A and
a probe current of 1.50 nA. Acquisition time was set
at 50 s and spectra were processed by ZAF corrections
and quantified using natural mineral standards. Mineral
chemical compositions were recalculated to mineral
Acknowledgements
southern Barberton terrain are at odds with the widely
established models of elevated heat production, heat
flow and a generally hotter and weaker crustal environment in the Archaean. The style of metamorphism in
the high-grade granitoid-gneiss terrain suggests that the
Mesoarchaean crust was sufficiently cold and rigid to
undergo tectonic stacking and support crustal overthickening. The structural integrity and absence of pervasive
ductile fabrics in this terrain confirms that it behaved as
a cohesive, rigid block during D2 burial and exhumation. Age constraints on the timing of peak metamorphism and cessation of D2 tectonism combined with the
depths of burial of this terrain suggest that it was tectonically exhumed at rates on the order of 2–5 mm/a,
which are comparable to the rates of exhumation in modern orogenic environments. In conclusion, the results of
this study suggest that the ca. 3.23 Ga amalgamation
of the Barberton granitoid-greenstone terrain occurred
by rapid, lateral plate-tectonic processes that involved
crustal blocks with a thermal structure and rheology
similar to the present day. This study also provides corroborating evidence that the first episode of cratonization
and the establishment of an insulating lithospheric mantle below the Kaapvaal Craton had already occurred by
ca. 3.23 Ga.
as a fundamental tectonic break between the southern
granitoid-gneiss terrain and the greenstone sequence in
the BGB. Kinematic indicators in the mylonites indicate
that the granitoid-gneiss terrain was uplifted relative to
the greenstone belt during deformation and Kisters et
al. (2003) suggested that the kinematics of the highly
strained granite-greenstone contacts and juxtaposition of
the high- and low-grade metamorphic terrains along this
contact indicate the presence of an extensional detachment. The main fabric-forming event in the Theespruit
Formation occurred subsequent to peak metamorphism
and was initiated under high-grade amphibolite-facies
conditions and continued during retrogression into the
greenschist-facies. The combination of near-isothermal
decompression and pervasive and ongoing extensional
shearing during the retrogression of the TSB indicates
that the fabric in these rocks is, indeed, an exhumation
fabric. Details of the exhumation process, including a
discussion on the late-stage steepening of fabrics, is presented in Kisters et al. (2003) and will, therefore, not be
repeated here.
The detailed P–T and geochronological data from
the Theespruit Formation allow to constrain the rates of
exhumation along the extensional detachment. Dziggel
(2002) provide the most robust and precise age constraint
for peak metamorphism in the southern granitoid-gneiss
terrain of 3229 ± 4 Ma. De Ronde and Kamo (2000) suggested that the D2 deformation in the BGB was a ca.
3 Ma, short-lived collisional event between 3229 and
3226 Ma. The post-tectonic, 3215 Ma Dalmein pluton
(Kamo and Davis, 1994) constrains the exhumation of
the high-grade rocks to have occurred within ca. 15 Ma.
Considering the depths of burial, these time constraints
point to average exhumation rates of ca. 1.5 mm/a, if the
intrusion of the Dalmein pluton is taken as the upper
time constraint, or ca. 5 mm/a when using the 3 Ma age
constraint. We emphasize that these rates must be taken
as minimum estimates as errors on the timing of peak
metamorphism and the duration of the D2 deformation
according to De Ronde and Kamo (2000) overlap. Moreover, any slower exhumation rates would have led to
the thermal equilibration of the high-P, low-T rocks.
In general, however, these rates compare well with the
exhumation rates of crustal rocks in younger orogenic
belts subsequent to collisional tectonism and burial (e.g.
Abbott and Silver, 1997).
6. Conclusions
The extremely low apparent geothermal gradients
that are consistently reported from both supracrustal
and basement high-grade metamorphic sequences in the
Actual
53.07
0.00
29.93
0.00
0.00
12.01
4.53
0.47
100.01
2.40
1.60
0.58
0.40
0.03
5.01
0.59
0.41
Measured
−2.1
n/d
4.7
n/d
n/d
1.7
4.0
12.8
% Dev
Plagioclase
54.21
0.07
28.53
0.37
0.13
11.80
4.35
0.41
99.87
2.46
1.52
0.57
0.38
0.02
4.98
0.60
0.40
Actual
39.43
0.00
22.15
23.41
0.34
10.74
3.95
100.02
2.99
1.98
1.49
0.02
1.22
0.32
8.02
0.40
0.49
0.01
0.11
0.55
Measured
0.6
n/d
0.5
0.6
−73.5
0.4
−6.3
% Dev
Garnet
39.19
0.00
22.05
23.27
0.59
10.70
4.20
100.00
2.98
1.98
1.48
0.03
1.21
0.34
8.03
0.40
0.48
0.01
0.11
0.55
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Ti
Fe
Mn
Mg
K
Total
Mg#
Actuala
40.80
1.61
15.42
0.00
11.76
0.21
20.07
0.00
11.46
101.33
2.83
1.17
0.09
0.08
0.68
0.01
2.08
1.01
7.96
75.25
Measured
1.0
−14.6
2.3
n/d
4.9
80.1
−1.4
n/d
9.8
% Dev
Biotite
40.38
1.85
15.78
0.00
11.18
0.04
20.36
0.00
10.33
100.02
2.81
1.19
0.11
0.10
0.65
0.00
2.11
0.92
7.89
76.44
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Al IV
Al VI
Cr
Fe
Mg
Total
Mg#
34.84
0.00
20.97
1.14
3.84
0.00
38.86
0.03
0.00
0.00
99.67
2.88
1.11
0.93
0.07
0.27
4.79
10.06
94.75
Actuala
36.74
0.00
21.77
1.25
4.11
0.00
36.06
0.00
0.00
0.00
99.93
3.01
0.99
1.12
0.08
0.28
4.41
9.89
93.99
Measured
5.2
n/d
3.7
9.1
6.6
n/d
−7.8
−
n/d
n/d
% Dev
Chlorite
Table A.1
Comparison of actual published major element concentrations of mineral standards and measured values determined by SEM-ED analysis
SiO2
TiO2
Al2 O3
FeO
MgO
CaO
Na2 O
K2 O
Total
Si
Al
Ca
Na
K
Total
XAn
XAb
SiO2
TiO2
Al2 O3
FeO
MnO
MgO
CaO
Total
Si
Al
Fe
Mn
Mg
Ca
Total
XPy
XAlm
XSpss
XGrss
Fe/Fe + Mg
SiO2
TiO2
Al2 O3
Cr2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
Si
Ti
Al
Fe
Mn
Mg
Ca
Na
K
Total
Mg#
Actuala
42.21
4.99
14.15
0.00
11.96
0.00
12.16
10.22
1.86
2.35
99.90
6.09
0.54
2.41
1.44
0.00
2.62
1.58
0.52
0.43
15.64
64.44
Measured
4.4
5.4
−5.3
n/d
5.9
−
−5.3
−0.8
−39.8
12.8
% Dev
Hornblende
40.37
4.72
14.90
0.00
11.25
0.09
12.80
10.30
2.60
2.05
99.08
5.89
0.52
2.56
1.37
0.01
2.78
1.61
0.74
0.38
15.87
66.97
Structural formulae were calculated on the basis of 8 oxygens for plagioclase, 12 for garnet, 11 for biotite, 14 for chlorite and 23 for hornblende XPy, XAlm, XSpss and XGrss as defined by Spear
(1993). XAn = Ca/(Ca + Na), XAb = Na/(Ca + Na), Mg# = 100 × Mg/(Mg + Fe); n/d = not determined.
a Recalculated as water-free equivalent.
110
(A.1)
(A.2)
Chopin, C., 1984. Coesite and pure pyrope in high-grade blueschists of
the western Alps: a first record and some consequences. Contrib.
Mineral. Petrol. 86, 107–118.
Choukroune, P.H., Bouhallier, A., Arndt, N.T., 1995. Soft lithosphere
during periods of Archaean crustal growth or crustal reworking.
In: Coward, M.P., Ries, A.C. (Eds.), Early Precambrian Processes,
vol. 95. Geol. Soc. Lond. Spec. Publ., pp. 67–86.
Cloete, M., 1999. Aspects of volcanism and metamorphism of the
Onverwacht Group lavas in the southwestern portion of the Barberton Greenstone Belt, Memoirs of the Geological Survey of South
Africa, vol. 84, 232 pp.
Collins, W.J., Van Kradendonk, M.J., Teyssier, C., 1998. Partial convective overturn of the Archaean crust in the eastern part of the
Pilbara Craton, Western Australia: 2. Driving mechanisms and tectonic implications. J. Struct. Geol. 20, 1405–1424.
Condie, K.C., 1994. Greenstones through time. In: Condie, K.C. (Ed.),
Archaean Crustal Evolution. Elsevier, Amsterdam, pp. 85–120.
Cox, R.A., Wilton, D.H.C., Kosler, J., 2003. Laser ablation U–Th–Pb
in situ dating of zircon and allanite: an example from the October
Harbour Granite, Central coastal Labrador, Canada. Can. Mineral.
41, 273–291.
Dann, J.C., 2000. The 3.5 Ga Komati Formation, Barberton Greenstone
Belt, South Africa: part I. New maps and magmatic architecture.
S. Afr. J. Geol. 103, 47–68.
Davies, G.F., 1992. On the emergence of plate tectonics. Geology 20,
963–966.
Davies, G.F., 1995. Punctuated evolution of the earth. Earth Planet.
Sci. Lett. 136, 363–379.
Davis, G.A., 1980. Structural characteristics of metamorphic core complexes, southern Arizona. Geol. Soc. Am. Memoir 153, 35–77.
De Ronde, C.E.J., De Wit, M.J., 1994. The tectonic history of the Barberton greenstone belt, South Africa: 490 million years of Archaean
crustal evolution. Tectonics 13, 983–1005.
De Ronde, C.E.J., Kamo, S.L., 2000. An Archaean arc–arc collisional
event: a short-lived, (ca. 3 Myr) episode, Weltevreden area, Barberton Greenstone Belt, South Africa. J. Afr. Earth Sci. 30, 219–248.
De Wit, M.J., 1982. Gliding and overthrust nappe tectonics in the Barberton greenstone belt. J. Struct. Geol. 4, 117–136.
De Wit, M.J., 1998. On Archaean granites, greenstones, cratons and
tectonics: does the evidence demand a verdict? Precambrian Res.
91, 181–226.
De Wit, M.J., Armstrong, R.A., Hart, R.J., Wilson, A.H., 1987. Felsic
igneous rocks within the 3.3 to 3.5 Ga Barberton greenstone belt:
high-level equivalents of the surrounding tonalite trondhjemite terrain, emplaced during thrusting. Tectonics 6, 529–549.
De Wit, M.J., Fripp, R.E.P., Stanistreet, I.G., 1983. Tectonic and stratigraphic implications of new field observations along the southern
part of the Barberton greenstone belt, Special Publication. Geol.
Soc. S. Afr. 9, 21–29.
De Wit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., De Ronde,
C.E.J., Green, R.W.E., Tredoux, M., Peberdy, E., Hart, R.A., 1992.
Formation of an Archaean continent. Nature 357, 553–562.
De Wit, M.J., Tinker, J., 2004. Crustal structures across the central
Kaapvaal craton from deep-seismic reflection data. S. Afr. J. Geol.
107, 185–206.
Dunstan, L.P., Gramch, J.W., Barnes, I.L., Purdy, W.C., 1980. The
absolute abundance and the atomic weight of a reference sample
of thallium. J. Res. Natl. Bur. Stand. 85, 1–10.
Dziggel, A., 2002. The petrogenesis of ‘lower’ Onverwacht Group clastic metasediments and related metavolcanic rocks in the southern
part of the Barberton Mountain Land, South Africa. Ph.D. thesis,
University of the Witwatersrand, Johannesburg, 230 pp.
Complex on the island of Stjernøy in the Seiland
Province of northern Norway. This sample has a conventional TIMS U–Pb age of 520 ± 5 Ma (Pedersen et
al., 1988). During the span of this study, the LAC titanite
yielded a Concordia Age of 523 ± 5.2 Ma (N = 22).
A.4. Corrections for initial common Pb
(208 Pb/(D∗ + C)measured ) − (208 Pb∗ /D∗ )
(208 Pb/Ccommon ) − (208 Pb∗ /D∗ )
This study used the “208 method”, based on assumption that the ratio of 232 Th to the parent U isotope in the
analyzed sample has not been disturbed following the
closure of U–Pb and Th–Pb isotopic systems and that
any excess 208 Pb (i.e., 208 Pb–208 Pb* ) can be attributed
to the common Pb component (see Kosler et al., 2002).
The Pb isotopic compositions for common Pb used in this
study are from the model of Stacey and Kramers (1975).
This approach requires the assumption of known age and
concordant composition in the U–Th–Pb system that are
used to calculate the radiogenic 208 Pb* /Daughter* ratios.
If C is the contribution of common-Pb to the daughter
(D*) radiogenic Pb signal, the correction equation has
the form:
f =
(208 Pb/206 Pbmeasured ) − (208 Pb∗ /206 Pb∗ )
(208 Pb/206 Pbcommon ) − (208 Pb∗ /206 Pb∗ )
As an example, Eq. (A.1) for 206 Pb becomes:
f =
References
Abbott, L.D., Silver, E.A., 1997. Measurements of tectonic surface
uplift in a young collisional mountain belt. Nature 385, 501–507.
Anhaeusser, C.R., 1973. The evolution of the Precambrian crust of
southern Africa. Philos. Trans. R. Soc., Lond. A 2731, 359–388.
Anhaeusser, C.R., 1984. Structural elements of Archaean granitegreenstone terranes as exemplified by the Barberton Mountain
Land, southern Africa. In: Kröner, A., Greiling, R. (Eds.), Precambrian Tectonics Illustrated. Stuttgart, 418 pp.
Anhaeusser, C.R., Robb, L.J., Viljoen, M.J., 1981. Provisional geological map of the Barberton Greenstone Belt and surrounding granitic
terrane, Eastern Transvaal and Swaziland, Geological Society of
South Africa, scale 1:250 000.
Armstrong, R.A., Compston, W., De Wit, M.J., Williams, I.S., 1990.
The stratigraphy of the Barberton Greenstone Belt revisited: a
single zircon ion microprobe study. Earth Planet. Sci. Lett. 101,
90–106.
Chardon, D., Peucat, J.J., Jayananda, M., Choukroune, P., Fanning,
C.M., 2002. Archaean granite-greenstone tectonics at Kolar, (South
India): Interplay of diapirism and bulk inhomogeneous contraction
during juvenile magmatic accretion. Tectonics 21, 700–717.
Cherniak, D.J., 1993. Lead diffusion in titanite and preliminary results
on the effects of radiation damage on Pb transport. Chem. Geol.
110, 177–194.
109
111
studies—a comparison of laser ablation ICPMS and SIMS techniques. Chem. Geol. 182, 605–618.
Kosler, J., Sylvester, P.J., 2003. Present trends and the future of zircon in
geochronology: Laser ablation ICP-MS. Rev. Mineral. Geochem.
V 53, 243–275.
Kretz, R., 1983. Symbols for rock-forming minerals. Am. Mineral. 68,
277–279.
Kröner, A., Byerly, G.R., Lowe, D.R., 1991. Chronology of early
Archaean granite-greenstone evolution in the Barberton Mountain
land, South Africa, based on precise dating by single zircon evaporation. Earth Planet. Sci. Lett. 103, 41–54.
Kröner, A., Hegner, E., Wendt, J.I., Byerly, G.R., 1996. The oldest
part of the Barberton granitoid-greenstone terrain, South Africa,
evidence for crust formation between 3.5 and 3.7 Ga. Precambrian
Res. 78, 105–124.
Kusky, T.M., Polat, A., 1999. Growth of granite-greenstone terranes
at convergent margins, and stabilization of Archaean cratons.
Tectonophysics 305, 43–73.
Lister, G.S., Davis, G.A., 1989. The origin of metamorphic core complexes and detachment faults formed during Tertiary continental
extension in the northern Colorado River region, U.S.A. J. Struct.
Geol. 11, 65–94.
Lopez-Martinez, M., York, D., Hanes, J.A., 1992. A 40 Ar/39 Ar
geochronological study of komatiites and komatiitic basalts from
the lower Onverwacht volcanics: Barberton Mountain Land, South
Africa. Precambrian Res. 57, 91–119.
Lowe, D.R., 1994. Accretionary history of the Archaean Barberton
Greenstone Belt (3.55–3.22 Ga), southern Africa. Geology 22,
1099–1102.
Lowe, D.R., 1999. Geologic evolution of the Barberton Greenstone
Belt and vicinity. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic
Evolution of the Barberton Greenstone Belt, 329. Special Paper
of the Geological Society of America, South Africa, pp. 287–
312.
Lowe, D.R., Byerly, G.R., Ransom, B.L., Nocita, B.R., 1985. Stratigraphic and sedimentological evidence bearing on structural repetition in Early Archaean rocks of the Barberton Greenstone Belt,
South Africa. Precambrian Res. 27, 165–186.
Lowe, D.R., Byerly, G.R., Heubeck, C., 1999. Structural divisions and
development of the west-central part of the Barberton Greenstone
Belt. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic Evolution of the
Barberton Greenstone Belt, 329. Special Paper of the Geological
Society of America, South Africa, pp. 37–82.
Ludwig, K.R., 1999. Isoplot/Ex v. 2.6. Berkeley Geochronological
Center Special Publication no. 1a.
Marshak, S., 1999. Deformation style way back when: thoughts on the
contrasts between Archaean/Paleoproterozoic and contemporary
orogens. J. Struct. Geol. 21, 1175–1182.
Miyashiro, A., 1961. Evolution of metamorphic belts. J. Petrol. 2,
277–311.
Passchier, C.W., Trouw, R.A.J., 1996. Microtectonics. SpringerVerlag, Berlin.
Pedersen, R.-B., Dunning, G., Robins, B., 1988. U–Pb ages of
nepheline syenite pegmatites from the Seiland Province, N. Norway. In: Gayer, R.A. (Ed.), The Caledonide Geology of Scanidinavia, Graham and Trotman, 381 pp., pp. 3–8.
Platt, J.P., Vissers, R.L.M., 1980. Extensional structures in anisotropic
rocks. J. Struct. Geol. 2, 397–410.
Poli, S., 1993. The amphibolite-eclogite transformation: an experimental study on basalt. Am. J. Sci. 293, 1061–1107.
Rawlings-Hinchey, A.M., Sylvester, P.J., Myers, J.S., Dunning, G.R.,
Kosler, J., 2003. Paleoproterozoic crustal genesis: calc-alkaline
R.A., 2002. Metamorphism of the granite-greenstone terrane
south of the Barberton greenstone belt, South Africa: an insight
into the tectono-thermal evolution of the ‘lower’ portions of the
Ernst, W.G., 1973. Blueschist metamorphism and P–T regimes in active
subduction zones. Tectonophysics 17, 255–272.
Ernst, W.G., 1975. Systematics of large-scale tectonics and age progressions in Alpine and circum-Pacific blueschist belts. Tectonophysics 26, 229–246.
Ernst, W.G., 1988. Tectonic history of subduction zones inferred from
retrograde blueschist P–T paths. Geology 16, 1081–1084.
Ferry, J.M., 1984. A biotite isograd in south-central Maine, USA:
mineral reactions, fluid transfer and heat transfer. J. Petrol. 25,
871–893.
Fonneland, H.C., Trond, L., Martinsen, O.J., Pedersen, R.B., Kosler,
J., 2004. Detrital zircon ages: a key to understanding the deposition
of deep marine sandstones in the Norwegian Sea. Sediment. Geol.
164, 147–159.
Fripp, R.E.P., Van Nierop, D.A., Callow, M.J., Lilly, P.A., Du Plessis,
L.U., 1980. Deformation in part of the Archaean Kaapvaal craton,
South Africa. Precambrian Res. 13, 241–251.
Frost, B.R., Chamberlain, K.R., Schumacher, J.C., 2000. Sphene,
(titanite): phase relations and role as a geochronometer. Chem.
Geol. 172, 131–148.
Hamilton, W.B., 1998. Archaean magmatism and deformation were
Hodych, J.P., Cox, R.A., Kosler, J., 2004. An equatorial Laurentia at
550 Ma confirmed by Grenvillian inherited zircons dated by LAM
ICP-MS in the Skinner Cove volcanics of western Newfoundland:
implications for inertial interchange true polar wander. Precambrian Res. 129, 93–113.
Holdaway, M.J., 1971. Stability of andalusite and the aluminum phase
diagram. Am. J. Sci. 271, 97–131.
Holland, T.J.B., Powell, R., 1998. An internally consistent thermodynamic data set for phases of petrological interest. J. Metamorph.
Geol. 16, 309–343.
Horn, I., Hinton, R.W., Jackson, S.E., Longerich, H.P., 1997. Ultratrace element analysis of NIST SRM 616 using laser ablation microprobe-inductively coupled plasma–mass spectrometry
(LAM–ICP-MS): a comparison with secondary ion mass spectrometry, (SIMS). Geostand. Newslett. 21, 191–203.
Horn, I., Rudnick, R.L., McDonough, W.F., 2000. Precise elemental
and isotope ratio determination by simultaneous solution nebulization and laser ablation-ICP-MS: application to U–Pb geochronology. Chem. Geol. 164, 281–301.
Hoschek, G., 2001. Thermobarometry of metasediments and metabasites from the eclogite zone of the Hohe Tauern, Eastern Alps,
Austria. Lithos 59, 127–150.
Jackson, S.E., Longerich, H.P., Horn, I., Dunning, G.R., 1996. The
application of laser ablation microprobe (LAM)-ICP-MS to in situ
U–Pb zircon geochronology. J. Conf. Abstr. 1, 283.
Kamo, S.L., Davis, D.W., 1994. Reassessment of Archaean crustal
development in the Barberton Mountain Land, South Africa, based
on U–Pb dating. Tectonics 13, 167–192.
Kisters, A.F.M., Stevens, G., Dziggel, A., Armstrong, R.A., 2003.
Extensional detachment faulting and core-complex formation in
the southern Barberton granite-greenstone terrain, South Africa:
evidence for a 3.2 Ga orogenic collapse. Precambrian Res. 127,
355–378.
Kosler, J., Fronneland, H., Sylvester, P., Tubrett, M., Pedersen, R.B., 2002. U-Pb dating of detrital zircons for sediment provenance
112
Van Kranendonk, M.J., Hickman, A.H., Smithies, R.H., Nelson, D.N.,
Pike, G., 2002. Geology and tectonic evolution of the Archaean
North Pilbara terrain, Pilbara Craton, Western Australia. Econ.
Geol. 97, 695–732.
Van Vuuren, C.F., Cloete, M., 1995. A preliminary P–T–t path for the
emplacement of the Theespruit pluton, Barberton greenstone belt,
Centennial Geocongress Abstract Volume. Geol. Soc. S. Afr., pp.
315–318.
Verts, L.A., Chamberlain, K.R., Frost, C.D., 1996. U–Pb dating of
metamorphism: the importance of sphene growth in the contact aureole of the Red Mountain pluton, Laramie Mountains,
Wyoming. Contrib. Mineral. Petrol. 125, 186–199.
Viljoen, M.J., Viljoen, R.P., 1969a. An introduction to the geology
of the Barberton granite-greenstone terrain, Special Publication.
Geol. Soc. S. Afr. 2, 9–28.
Viljoen, M.J., Viljoen, R.P., 1969b. The geology and geochemistry of
the lower ultramafic unit of the Onverwacht Group and a proposed
new class of igneous rocks, Special Publication. Geol. Soc. S. Afr.
2, 55–86.
Will, T.M., Powell, R., Holland, T.J.B., 1990. A calculated petrogenetic grid for ultramafic rocks at low pressures in the system CaO–FeO–MgO–Al2 O3 –SiO2 –CO2 –H2 O. Contrib. Mineral.
Petrol. 105, 347–358.
Williams, P.R., Currie, K.L., 1993. Character and regional implications
of the sheared Archaean granite-greenstone contact near Leonora,
Western Australia. Precambrian Res. 62, 343–365.
Williams, D.A.C., Furnell, R.G., 1979. Reassessment of part of the
Barberton type area, South Africa. Precambrian Res. 9, 325–
347.
Xie, X., Byerly, G.R., Ferrell, R.E., 1997. IIb trioctahedral chlorite from the Barberton Greenstone Belt: crystal structure and
rock composition constraints with implications to geothermometry. Contrib. Mineral. Petrol. 126, 275–291.
Yearron, L.M., 2003. Archaean granite petrogenesis and implications
for the evolution of the Barberton Mountain Land, South Africa.
Ph.D. thesis, Kingston University, 315 pp.
magmatism of the Torngat Orogen, Voisey’s Bay area, Labrador.
Precambrian Res. 125, 55–85.
SACS (South African Committee for Stratigraphy), 1980. Stratigraphy
of South Africa, Part 1 (Compiled by L.E. Kent), Lithostratigraphy
of the Republic of South Africa, South West Africa/Namibia and
the Republics of Bophutatswana, Transkei and Venda: Handbook
of the Geological Survey of South Africa, vol. 8, 690 pp.
Sandiford, M., McLaren, S., 2002. Tectonic feedback and the ordering of heat producing elements within the continental lithosphere.
Earth Planet. Sci. Lett. 204, 133–150.
Scott, D.J., St. Onge, M.R., 1995. Constraints on Pb closure temperature in titanite based on rocks from the Ungava Orogen, Canada:
implications for U–Pb geochronology and P-T-t path determinations. Geology 23, 1123–1126.
Shirey, S.B., Harris, J.W., Richardson, S.H., Fouch, M., James, D.E.,
Cartigny, P., Deines, P., Viljoen, F., 2003. Regional patterns in the
paragenesis and age of inclusions in diamond, diamond composition, and the lithospheric seismic structure of Southern Africa.
Lithos 71, 243–258.
Smith, D.C., 1984. Coesite in clinopyroxene in the Caledonides and
its implications for geodynamics. Nature 310, 641–644.
Spear, F.S., 1993. Metamorphic Phase Equilibria and Pressure–
Temperature–Time Paths. Mineralogical Society of America
Monograph, Washington, DC, 799 pp.
Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead
isotope evolution by a two stage model. Earth Planet. Sci. Lett. 26,
207–221.
Stevens, G., Droop, G.T.R., Armstrong, R.A., Anhaeusser, C.R., 2002.
Amphibolite facies metamorphism in the Schapenburg schist belt:
A record of the mid-crustal response to ∼3.23 Ga terrane accretion
in the Barberton greenstone belt. S. Afr. J. Geol. 105, 271–284.
Tiepolo, M., 2003. In situ Pb geochronology of zircon with laser
ablation–inductively coupled plasma–sector field mass spectrometry. Chem. Geol. 199, 159–177.
Van Kradendonk, M.J., 2004. Preface: Archaean tectonics 2004: a
review. Precambrian Res. 131, 143–151.
a
Barberton (South Africa) TTG magmas: Geochemical and
experimental constraints on source-rock petrology,
pressure of formation and tectonic setting
J.D. Clemens a,∗ , L.M. Yearron a , G. Stevens b
Received 11 August 2004; received in revised form 20 July 2006; accepted 11 August 2006
School of Earth Sciences and Geography, CEESR, Kingston University, Penrhyn Rd, Kingston-upon-Thames, Surrey KT1 2EE, UK
b Department of Geology, Stellenbosch University, Matieland 7602, South Africa
Abstract
In the southern area of the Barberton Mountain Land, TTG magmas were produced during two distinct, major, magmatic events at
ca. 3.44 and 3.23 Ga. Here, as in many Archaean terranes, tonalite-trondhjemite-granodiorite (TTG) plutons are closely associated
with basaltic to komatiitic greenstones, in this case with some metamorphosed to amphibolite facies, suggesting a possible genetic
connection between the two rock groups. Previous partial melting experiments on metabasic rocks have shown that tonalitic to
trondhjemitic melts can be produced, coexisting with amphibole- and plagioclase-rich restites, at pressures of 0.8–1.5 GPa, and
garnet- and pyroxene-rich restites at P ≥ 1.5 GPa. The present experiments on a Barberton greenstone amphibolite confirm the
higher pressure findings, except that some amphibole was probably still present. In the case of the 3.23 Ga plutons, the inferred
geotherm is consistent with that obtained from metamorphic assemblages of this age from within the Theespruit Formation of the
Onverwacht Group. The commonly scattered major- and trace-element variations in the Barberton TTG suite imply that magmatic
crystal fractionation played a subordinate role in producing the geochemical variations of the magmas. The different TTG plutons
probably represent separate magma batches, and the scattered trends within the plutons probably reflect heterogeneities within their
source-rocks. The ␧Nd values suggest that the TTGs were derived from juvenile crustal sources with depleted-mantle signatures.
Thus, metabasaltic rocks are the likely sources of the TTG magmas. However, our partial melting experiments on a typical Lower
Onverwacht greenstone amphibolite appear to rule out these particular rocks as sources of the local TTG magmas. Instead, it seems
likely that possibly more ancient, less potassic, high-grade, metabasic rocks were the sources of the TTG magmas. Trace-element
modelling shows that the TTG suite could have been derived through partial melting of primitive basaltic sources, producing
plagioclase-free, hornblende-bearing granulitic to eclogitic restites with >30% garnet. Experiments on garnet stability in the nearliquidus mineral assemblage of a typical 3.44 Ga Barberton trondhjemite constrain magma generation to a pressure of at least
1.47 GPa. This suggests that the Barberton crust was relatively cool and at least 50 km thick by 3.44 Ga. The same general argument
of high-P melting would hold for the ca. 3.2 Ga trondhjemites and tonalities, although the minimum P of melting has not been
determined for these rocks. In the case of these rocks, believed to have formed in response to a major terrane accretion event, the highP–moderate-T signature is also indicated by recent metamorphic studies. In contrast, information is scarce on the processes operating
during the 3.44 Ga magmato-metamorphic event. The P–T conditions during the 3.44 Ga event imply an apparent geothermal gradient
of <20 ◦ C/km. The transport of fertile, hydrated metabasic material to such depths suggests that both the 3.23 and 3.44 Ga magmatic
events resulted from significant and rapid crustal thickening. This, in turn, suggests that compressional tectonics operated in the
Barberton greenstone belt prior to 3.44 Ga.
Corresponding author. Fax: +44 20 8547 7497.
E-mail addresses: [email protected] (J.D. Clemens), l [email protected] (L.M. Yearron), [email protected] (G. Stevens).
Keywords: Barberton; Granite; TTG; Archaean; Crustal thickening
∗
doi:10.1016/j.precamres.2006.08.001
54
1. Introduction
collisional thickening of oceanic or arc crust). There are
also some difficulties in explaining how subducting slabs
retain the required high H2 O contents to the depths necessary for melting and how wet slab melts can ascend
through the mantle wedge without being consumed by
reactions with the peridotite (see, e.g. Rapp et al., 1999;
Prouteau et al., 2001). The fluid-present experiments of
Prouteau et al. (1999), on a dacite with TTG-like geochemistry, showed that plagioclase fractionation could
only be avoided for melt H2 O contents of 10 wt% or
more. This work also showed that the necessary garnet
could not have been present near the liquidus of such a
magma, at near-source P, T and these fluid conditions.
Fluid-absent melting was rejected because such melts,
formed at T ≤ 900 ◦ C, would be too silicic and potassic, and their interaction with the mantle wedge would
not alter these parameters significantly. However, the
chemistry of the fluid-absent partial melts would depend
strongly on the composition of the protolith (Moyen and
Stevens, 2005). Also, this neglects the possibility that the
slab might not be the setting for TTG genesis and that
melting could be at much higher T. The experiments of
Rapp et al. (1999) are also instructive. These showed that,
with melt:mantle peridotite ratios around 2, slab-derived
melts would survive reaction with the mantle. However,
for ratios near 1, the melts would be entirely consumed by
reaction with the peridotite. Adakites (supposed modern
slab melts) have MgO contents and Mg#s that suggest
interaction with the rocks of the mantle wedge. However, TTGs have lower MgO and Mg#s than adakitic
rocks, and adakitic intrusive rocks do occur in settings
unrelated to subduction (e.g. Xu et al., 2002). This suggests that TTG magmas may not have interacted with
mantle rocks, and that therefore they were generated in
settings in which the magmas could reach emplacement
levels without travelling through the mantle (i.e. not by
slab melting). Martin and Moyen (2002) and Martin et
al. (2005) showed that TTGs generally have higher Mg#
than experimentally produced partial melts of basaltic
rocks. They also point out that Mg# in parental TTG
magmas increased over the Archaean, from 4.0 to 2.5 Ga,
suggesting that the degree of interaction between felsic
melts and mantle peridotite increased over time. Smithies
(2000) showed that such interaction must have been very
weak or even absent in TTG magmas older than 3 Ga,
and in around 50% of the post-3 Ga TTG magmas as
well. Using a similar dataset, Martin et al. (2005) concluded that mantle wedge was either thin or non-existent
before 3.4 Ga but that there was a general steepening of
subduction angle (and thickening of the mantle wedge)
at later times. Despite the volume of work carried out on
the problem, the precise nature of the melting reactions
J.D. Clemens et al. / Precambrian Research 151 (2006) 53–78
Tonalite-trondhjemite-granodiorite rocks typically
form about two-thirds of the presently accessible
Archaean crust (Jahn et al., 1984). The onset of this magmatism is generally believed to represent the transition
from dominantly mafic crust, to crust with a significant
felsic component (Glikson, 1979). Crustal stabilisation
and cratonisation are believed to have developed in short,
intense episodes of continental growth, involving magmatic accretion (e.g. Wells, 1981) as well as tectonic
thickening and high-grade metamorphism (e.g. De Wit,
1998). As a result, the TTG rocks form an essential element in the ‘protocontinental’ stage of crustal evolution
(Barker, 1979).
The origin of tonalite-trondhjemite magmas has been
widely debated. Various suggestions have included fractional crystallisation of basaltic melts (e.g. Arth et al.,
1978), partial melting of mantle rocks (e.g. Moorbath,
1975), and the partial melting of pre-existing tonalites
(e.g. Johnston and Wyllie, 1988). However, the most
widely accepted mechanism for the origin of TTG magmas is by partial melting of hydrous metabasaltic rocks,
i.e. greenstones, amphibolites and eclogites, under a variety of fluid conditions and in a variety of tectonic settings
(e.g. Martin, 1987; Winther, 1996; Condie, 2005). This
latter category of petrogenetic models is largely based
on the fact that the chondrite-normalised REE patterns
of TTG rocks are typically HREE-depleted and LREEenriched. Since garnet readily accommodates HREEs, its
presence in the crystalline residuum may well account
for the HREE-depleted pattern (e.g. Jahn et al., 1981;
Rapp et al., 1991; Springer and Seck, 1997). Within
this group of models, the main competition is between
those that involve fluid-present (but usually highly H2 Odeficient) melting of altered mafic rocks in the downgoing slab (e.g. Prouteau et al., 1999) and those advocating higher-temperature, fluid-absent melting of similar
materials, mainly in the deep thickened crust, in a variety of tectonic settings (e.g. Rapp et al., 2003). Foley
et al. (2002) presented geochemical evidence that melts
with some trace-element ratios similar to those of TTG
rocks can be produced by partial melting of low-Mg garnet amphibolites, but not by partial melting of eclogites.
They concluded that this melting must have taken place
in subduction zones. However, with this model, there
are residual difficulties in reproducing some important
trace-element characteristics of TTGs (e.g. their Sr, U
and Th concentrations). Whatever the source (garnet
amphibolite or eclogite), the high pressures necessary
to generate TTG melts could still be produced in settings other than subduction zones (e.g. post-subduction
55
(1) The 3.50–3.30 Ga Onverwacht Group, composed
largely of mafic and ultramafic volcanic rocks, with
minor units of felsic volcanic and volcaniclastic
rocks, as well as sediments.
The Palaeo- to Meso-archaean Barberton greenstone belt consists of a well-preserved, early Archaean
(3.5–3.2 Ga) volcano-sedimentary succession, comprised of three major lithostratigraphic units. From the
base upward, these are:
2. Geological setting of TTG magmatism
associated with the Barberton greenstone belt
ated with basaltic to komatiitic greenstones. The Barberton TTG plutons and the highest-grade, amphibolitefacies greenstone rocks are juxtaposed against each other
(Fig. 1), and there are many greenstone remnants within
the TTG plutons. As a result, a number of South African
geologists have concluded that the greenstone rocks represent the source of the TTG magmas (e.g. Robb and
Anhaeusser, 1983; Robb, 1983). In this paper we investigate the Barberton TTG rocks, using geochemical and
experimental approaches, to address the nature of the
protolith and the conditions of TTG magma genesis.
Fig. 1. Summary of the geology of the southern Barberton region.
that produced TTG magmas, and the tectonic settings in
which this occurred, remain matters of debate.
Most granitic magmas, including the TTGs, were
initially markedly H2 O-undersaturated (e.g. Clemens,
1984; Scaillet et al., 1998; Prouteau et al., 1999).
Clemens and Watkins (2001) showed that the observed
systematic negative correlation between initial magma
temperature and melt H2 O content is consistent only
with the magmas being derived through fluid-absent partial melting of pre-existing, hydrous crustal rocks (either
deep in the crust or in the upper mantle). Melts with
tonalitic and trondhjemitic major-element compositions
have been produced by the fluid-absent partial melting
of metabasaltic rocks under a wide variety of conditions.
Clemens (2005) provides a review of all this experimental work. Rapp et al. (1991) presented a fairly comprehensive study in which they partially melted four natural
olivine-normative amphibolites in the pressure range of
0.8–3.2 GPa, at temperatures between 900 and 1150 ◦ C.
Results showed that tonalitic to trondhjemitic melts were
produced, coexisting with amphibole- and plagioclaserich restites, at pressures of 0.8 GPa, and garnet- and
pyroxene-rich restites at pressures ≥1.6 GPa.
In the Barberton Mountain Land, as in many
Archaean terrains, the TTG plutons are closely associ-
56
Armstrong et al., 1990; Kröner et al., 1996, 1991) has
demonstrated that there are three clear magmatic age
clusters within the TTG suite; the 3509 ± 8 Ma Steynsdorp pluton; the 3460 ± 5 to 3443 ± 4 Ma Stolzburg,
Theespruit and Doornhoek plutons; and the 3236 ± 1
to 3227 ± 1 Kaap Valley and Nelshoogte plutons. The
age of the youngest Kaap Valley–Nelshoogte TTG
generation coincides with the proposed age for the
major terrane accretion episode that assembled the
rocks of the greenstone belt (Kamo and Davis, 1994;
de Ronde and Kamo, 2000). The age of intrusion of
these magmas also coincides with the age of peak
high-pressure, amphibolite-facies metamorphism, documented from greenstone remnants within the ∼3450 Ma
TTG intrusions (Dziggel et al., 2002) and within the
Theespruit Formation of the greenstone belt (Kisters
et al., 2003). Thus, the TTG bodies have been subject to amphibolite-facies metamorphic conditions, and
the hornblende compositions and zoning in the plagioclase may have been slightly affected in the older
TTG generations. However, since the igneous mineral
assemblages are similar to those stable in the upper
amphibolite facies, there has been little effect of the
metamorphism on the mineralogy or chemistry of the
rocks. The metamorphism caused neither dehydration
nor partial melting in the TTG rocks, and was essentially
isochemical.
The tectonic setting and associations of the older
generations of TTGs is more difficult to constrain, principally because of the younger, high-grade metamorphic
overprint on these plutonic rocks and the metamafic
xenoliths that they include. In several areas within the
∼3450 Ma TTG suite, intrusion breccias indicate that
these plutons formed as high-level bodies. This is supported by zircon ages, from some of the felsic volcaniclastic components of the Onverwacht Group, that are
identical to the crystallization ages of the Theespruit
pluton (Armstrong et al., 1990). In places, the plutons cut across the lithological layering of amphibolite facies rocks, indicating their possible association
with an older, high-grade metamorphism. However, the
details of this metamorphism, as well as its timing
relative to the intrusion, are yet to be investigated in
detail.
The end of TTG magmatism in the southern Barberton granite-greenstone terrane is marked by the
intrusion of the post-tectonic, granodioritic Dalmein
pluton at 3216 ± 2 Ma (Kamo and Davis, 1994).
This also marks the first appearance of the more
potassic granodiorite-monzogranite-syenite suite that
is dominated by voluminous plutonism dated at
3107 Ma.
(2) The 3.26–3.22 Ga predominantly argillaceous Fig
Tree Group, comprised of a succession of
greywackes, cherts and shales, plus some dacitic
lavas and fragmental volcanic rocks.
(3) The ∼3.2 Ga arenaceous sedimentary Moodies
Group, which consists largely of feldspathic and
quartzose sandstones, polymictic conglomerates,
lesser siltstones and shales, and thin units of basalt,
jaspilite and magnetite-bearing shale.
This volcano-sedimentary sequence was intruded
by two main suites of granitoid magmas between ca.
3.50 and 3.10 Ga. The resulting Barberton granitoidgreenstone terrane was assembled during several
tectonomagmatic episodes between ∼3.5 and 3.1 Ga
(e.g. Anhaeusser and Robb, 1983; Robb and Anhaeusser,
1983; Armstrong et al., 1990; Kamo and Davis, 1994;
de Ronde and De Wit, 1994). Early ∼3.5 to 3.2 Ga plutonic suites are characterized by tonalites, trondhjemites
and granodiorites. The trondhjemites and granodiorites
are dominated by sodic plagioclase and quartz with
biotite as the major mafic mineral. The tonalites are
similar but hornblende is present in addition to biotite
and can even dominate the mafic mineral assemblage.
In all these rocks, the plagioclase crystals are euhedral
to subhedral and were evidently the first felsic phase to
crystallise, with the biotite and/or hornblende appearing later in the crystallisation sequence. The composite
and commonly internally heterogeneous plutons commonly possess internal contact relationships between
magmatic fractions. The plutons are relatively small
(<100 to ∼500 km2 ) and have largely concordant contacts with the greenstones. Some of the felsic rocks
are gneissose, particularly near their margins. These
structural features have been explained through either
the diapiric ascent and emplacement of the TTGs (e.g.
Viljoen and Viljoen, 1969; Anhaeusser and Robb, 1983),
the synkinematic, shallow crustal underplating of the
TTG suite at the base of the largely allochthonous and
thrusted greenstone sequences (e.g. De Wit et al., 1987;
Armstrong et al., 1990) or as structurally reworked basement, commonly with tectonic rather than intrusive contacts between the greenstones and the TTGs (e.g. Dziggel
et al., 2002; Kisters et al., 2003). The TTG rocks themselves generally contain very few mafic magmatic inclusions (enclaves), and the plagioclase feldspars do not
show textural evidence of resorbtion. These features suggest that magma mixing and mingling played little part in
the production of the TTG magmas, at least at emplacement level.
Collectively, zircon geochronology from several studies (Kamo and Davis, 1994; de Ronde and Kamo, 2000;
3. Geochemistry
3.1. Methods
57
Fig. 2. (a) Classification of the TTG rocks using normative anorthite
(an), albite (ab) and orthoclase (or), with fields defined by Barker
(1979). Circles, squares: present work; triangles: Anhaeusser and
Robb (1983). (b) A/CNK–SiO2 plot for the TTG rocks. A/CNK = mol
Al2 O3 /(CaO + Na2 O + K2 O). (c) K2 O–SiO2 plot showing the fields
defined by Le Maitre et al. (1989). Circles, squares: present work,
triangles: Anhaeusser and Robb (1983).
Fresh, unaltered rock samples collected during fieldwork were crushed and powdered using a jaw crusher,
roller mill and Tema mill. Major- and trace-element
data were obtained by X-ray fluorescence spectroscopy
(XRF), using a Philips 1404 spectrometer, at the University of Stellenbosch. The spectrometer is fitted with
a Rh tube and six analyzing crystals of LIF200, LIF220,
LIF420, PE, TLAP and PX1. The detectors used a gasflow proportional counter and a scintillation detector.
The gas-flow proportional counter uses P10 gas. Major
elements were analysed on fused glass beads at 50 kV
and 50 mA, and trace elements were analysed on pressed
powder pellets at 60 kV and 40 mA. Matrix effects were
corrected for by applying theoretical alpha factors and
measured line overlap factors to the raw intensities, with
the SuperQ Philips software. Standards used in the calibration were: AGV-1, BHVO-1, JG-1, JB-1, GSP-1,
SY-2, SY-3, STM-1, NIM-G, NIM-S, NIM-N, NIM-P,
NIM-D, BCR, GA, GH, DRN and BR. At this facility,
standard material AGV-1 is also routinely analysed as
a sample, to check for analytical error. Major elements
are typically within 1 rel.% of the standard values for
elements present at >10 wt%, within 2% for elements
present at concentrations between 1 and 10 wt%, and
within 7% for elements present at <1 wt%. Measured values for trace elements on AGV-1 were mostly within 10%
(usually 5%) of the accepted values, with the exception of
Y (20%). Cr and Ni values were affected by contamination from the Tema mill vessels used in sample grinding,
so results for these elements are omitted from the data set.
Rare-earth-element data were obtained by inductively
coupled plasma atomic emission spectroscopy (ICPAES) at the University of Stellenbosch and by inductively coupled plasma mass spectroscopy (ICP-MS) in
the NERC Facility at Kingston University, UK.
Most mineral and glass (quenched melt) analyses (in
the experimental run products to be described later) were
carried out on the JEOL 3200 SEM, fitted with an Oxford
Instruments ISIS EDS system, at Kingston University.
In the products of the TTG near-liquidus experiments,
glass areas were sufficiently large that good-quality analyses could be obtained by rastering the beam over large
patches of glass, to minimize counting losses on Na.
However, in the products of the partial melting experiments on greenstone amphibolite, glass areas were much
smaller. To obtain analyses essentially free from Na
counting losses, we used a LEO 140VP scanning electron microscope coupled to a Link ISIS energy dispersive
58
Selected trace-element compositions of the TTG suite
are displayed in Fig. 4, plotted as Harker diagrams.
Note that, for internal consistency and comparability,
all these plots use the XRF analyses and ICPMS REE
3.3. Trace elements
White, 1974), with the implication that they were derived
by partial melting of meta-igneous source rocks (Chappell and White, 1984). This suite is low- to medium-K,
calc-alkaline (Fig. 2c).
The Harker diagrams in Fig. 3 show the variations of
TiO2 , Al2 O3 , Na2 O, CaO, MgO and FeOT with silica
content. SiO2 contents range from ∼61 to 77 wt%. As
is typical for igneous suites, most oxides are negatively
correlated with SiO2 . The exceptions are K2 O (Fig. 2c)
and Na2 O, with scattered trends. These correlations are
generally displayed among the analyses from individual
plutons, as well as for the suite, as a whole. Despite the
trends displayed for the suite, there is a significant degree
of scatter. This point is discussed further below.
spectrometry system at the University of Stellenbosch.
The microscope was operated at 20 kV with a beam current of 120 nA and a probe current of 1.50 nA. Acquisition time was set at 50 s. Spectra were processed by ZAF
corrections and quantified using natural mineral standards. This instrument is fitted with a HEXLAND cryostage that allows samples to be cooled to near liquid-N2
temperature (∼−193 ◦ C), which effectively eliminates
Na analytical problems, even when analysing with a fully
focussed beam (see, e.g. Vielzeuf and Clemens, 1992).
3.2. Major elements
Major-, trace-element and REE data for the TTG suite
are presented in Appendix A in supplementary data.
Using the geochemical classification of Barker (1979),
the rocks of the TTG suite are classified as mainly trondhjemitic, with minor tonalitic or granitic components of
the plutons (Fig. 2a). Their A/CNK values (Fig. 2b) vary
between about 0.8 and 1.2. The presence of Hbl, Tit and
Mag define these rocks as I-type granites (Chappell and
Fig. 3. (a–f) Major-element Harker diagrams for the TTG-suite rocks with SiO2 > 60 wt%, analysed in the present study.
59
Fig. 5. Multi-element diagram for the TTG rocks, normalised to the
primitive mantle values of McDonough and Sun (1995). See text for
discussion.
dorp, Theespruit, Doornhoek, Batavia and Nelshoogte)
also show small depletions (negative anomalies) for Ba,
relative to Rb and Th. Compared with the other rocks
plotted here, sample BTV13A shows significantly higher
enrichments, across the spectrum.
Fig. 4. (a–f) Trace-element Harker diagrams for the TTG-suite rocks with SiO2 > 60 wt%, analysed in the present study.
analyses, presented in Appendix A in supplementary
data. For the isotope work (see below) the more accurate
isotope-dilution analyses for Sm and Nd are used. The
Harker plots typically show scattered distributions. Only
V forms a relatively “tight” (i.e. distinct) negative correlation with SiO2 (Fig. 4f). The data for some individual
plutons, however, do exhibit weak trends. However, as a
whole, the TTGs do not show tight trace-element trends
with SiO2 content. This suggests that crystal fractionation was not the dominant process in the formation of the
TTG suite. The main exception, the trend in V, is probably related to the crystallisation and fractionation of
oxide minerals (Fig. 4f). The causes of the geochemical
variation are discussed below.
The multi-element diagram of Fig. 5 shows trace
element variations normalised to the primitive mantle
values of McDonough and Sun (1995). As is common
for TTG-like rocks this plot shows considerable enrichment in LILEs and a negative Nb anomaly. Ti, Y and
Yb do not show significant enrichments, which is also
common in TTG suites. A number of plutons (Steyns-
60
from the Stolzburg pluton. The TiO2 and CaO and Zr
plots seem to suggest quite distinct trends for the rocks
of this pluton. Such tight trends (and the MgO trend in
Fig. 3b, for example) might be thought of as indicating
fractionation. However, if oxides such as Al2 O3 , MgO
and FeOT , are plotted even for genetically unrelated
metaluminous felsic rocks, the resulting trends are
similar. They probably reflect the stoichiometries of
the melting reactions that formed the magmas, and the
partitioning of elements between granitic (s.l.) liquids
and the residual solids. The melt compositions are quite
limited and the residual crystal phases will be similar,
especially for variable but generally similar source
rock compositions. This effectively buffers the majorelement contents of the melts and produces relatively
tight major-element trends. This is why the partial melts
of a vast range of crustal rock types are broadly granitic
in chemistry. The trace elements are not so constrained
because their concentrations are commonly not buffered
by a crystalline phase that contains the element as a
Fig. 6. (a–f) Selected major- and trace-element Harker plots for analyses of rocks from the Stolzburg pluton. See text for discussion.
3.4. Causes of geochemical variation
The plutons of the TTG suite are dominated by
the mineral assemblage Pl + Bt ± Hbl, with accessory
Ap + Aln + Fe–Ti oxide ± Tit. Crystallisation and fractionation of these minerals could explain the negative
trends displayed by CaO, Al2 O3 , MgO, FeOT , TiO2
and P2 O5 . Crystal fractionation trends are characterised
by quite tight inter-element correlation on Harker plots.
Good examples can be found in Wyborn et al. (2001),
which deals with differentiation of some mainly felsic
plutons in Australia, for which the field, mineralogical
and geochemical data are consistent with production of
the rock series by crystal fractionation of a single parent
magma. For the Barberton TTGs, however, there is generally more scatter in the major-oxide trends than would
be expected if crystal fractionation were the sole process
responsible for the variation.
As an example of this, Fig. 6 shows selected majorand trace-element Harker plots for analyses of rocks
61
Note that the Rb/Sr shows a flat trend that only kicks
upward at the very high-SiO2 end (Fig. 6e). Even then,
not all of the rocks with SiO2 >74 wt% form part of
this upward spike. Note also that the variation in Mg#
(Fig. 6f) shows only a rough negative correlation with
SiO2 and has a great degree of scatter (e.g. Mg# varying
between 27.67 and 50.23 at SiO2 contents close to
73 wt%). Again this is abnormal for variation controlled
by mixing or crystal fractionation processes.
We interpret these variations as most probably due to
initial variation among the magma fractions that formed
the TTG plutons, i.e. that the Stolzburg pluton was probably assembled by the aggregation of a number of different magma batches, each with a slightly different melt
composition. Evidence is accumulating that this is the
case for very many felsic intrusive bodies (e.g. Glazner
et al., 2004; Clemens et al., in press). The observed overall trends of increasing CaO, FeO and MgO contents,
major structural constituent. Zr is one of the exceptions
and, unsurprisingly, Zr trends are usually quite tight on
Harker plots. Another factor that contributes to scatter in
trace-element concentrations of melts is the apparently
common occurrence of disequilibrium during partial
melting (e.g. Bea, 1996). In the CaO plot for Stolzburg
(Fig. 6b), note that rocks with around 70–71 wt% SiO2 ,
have CaO contents varying from about <1 to >2.5 wt%.
This is a little more scatter than might be expected in
a series of rocks related by fractional crystallisation,
magma mixing or crystal unmixing. If plagioclase is
not present in the residual assemblage of the TTG
source rocks (as seems certain, for most), CaO will
be only weakly buffered, perhaps by clinopyroxene in
the melting residue, and scatter is expected. The Ba
plot (Fig. 6d) shows a large amount of scatter, with no
clear trend. This degree of scatter and lack of a trend
is also unusual for Ba in differentiated magmatic suites.
Fig. 7. REE patterns for the TTG plutons normalised to chondrite (Nakamura, 1974; Haskin et al., 1968, for Tb); (a) Batavia, Badplaas and
Rooihoogte, (b) Stolzburg and Theespruit, (c) Eerstehoek and Theeboom, (d) Kaap Valley and Nelshoogte, (e) Steynsdorp and Doornhoek.
62
best interpreted as indicating the absence of plagioclase
in the residual source. In addition, fractionation of a sufficient quantity of hornblende, from a felsic magma, to
overcome the effect of plagioclase fractionation, is not
feasible, on simple mass-balance grounds. The magmas
could not have contained sufficient ferromagnesian component and, in any case, the hornblende in the Barberton
TTG rocks is not an early-crystallising phase. Thus, the
rarity of negative Eu anomalies suggests that it is unlikely
that plagioclase fractionation occurred during the evolution of most Barberton TTG magmas. The somewhat
flatter REE patterns for Eerstehoek, Steynsdorp, Doornhoek and Kaap Valley may signify a lower abundance of
garnet in the restitic source of these magmas. Anhaeusser
and Robb (1983) analysed a number of TTG samples
from the same area but, for internal consistency, their
data are not plotted in Fig. 7. Nevertheless, these authors
record negative Eu anomalies (average Eu/Eu* values of
0.72–0.83) in a few samples from the Steynsdorp and
Doornhoek plutons, suggesting a degree of plagioclase
fractionation. This is compatible with petrographic and
textural evidence for the early crystallisation of plagioclase in the rocks.
Four of the analysed samples in our dataset (Appendix
A in supplementary data) have high K2 O/Na2 O (>1),
and thus could be considered not to belong to the TTG
suite. Batavia pluton sample BTV13A has a relatively
low SiO2 content (65.55 wt%) and K2 O/Na2 O = 1.21. It
is characterised by extremely elevated P2 O5 , Ba, Rb,
Sr, Y, Zr, Nb, Zn and REE but lacks a positive Eu
anomaly. These features, well portrayed in Fig. 5, are
consistent with a rock enriched in biotite, zircon, apatite
and Fe-Ti oxides. We interpret this as a cumulate, derived
from the TTG magma by magmatic segregation of these
phases; it is not plotted in Fig. 7. The remaining three
samples in this category (NLG9, STY4B and STZ23) all
have very high SiO2 contents (>75 wt%). The REE pattern for STY4B is shown as the dashed line in Fig. 7e).
This rock has a high SiO2 content (75.11 wt%), elevated
K2 O/Na2 O (1.28), relatively high Ba and low concentrations of Sr, Y and V. Its REE pattern is unexceptional,
but there is a shallow negative Eu anomaly. All of this
is consistent with an origin as a felsic differentiate of a
TTG magma; STZ23 is similar. However, sample NLG9
has extreme K2 O/Na2 O (4.86), high Rb, low Sr, high
REE and rather elevated LREE contents. It is also
strongly peraluminous (A/CNK = 2.88). These characteristics suggest that the NLG9 magma was probably
formed by partial melting of a minor metasedimentary
component within the TTG source region.
In summary, the generation of the majority of the
TTG magmas probably involved partial melting of mafic
with decreasing SiO2 and Na2 O are most probably due
to progressive partial melting, with incremental melt
extraction. However, the scattered geochemical variation
in the TTG rocks represents an overtone that is probably
due to local variations in the compositions of the sources
of individual magma batches. These separate batches
evidently failed to mix efficiently within the growing
plutons. Intraplutonic intrusive relationships between
different magma batches clearly demonstrate that
several of the TTG plutons are composite bodies. Such
geochemical and geological relationships are present in
the Badplaas, Theespruit and Nelshoogte plutons.
3.5. Rare earth elements
Fig. 7 shows the REE patterns for samples from
the TTG plutons. The samples are LREE-enriched and
HREE-depleted (relative to chondritic concentrations),
producing average (La/Yb)N values of 17–49 and YbN
values of 1–10. However, Eerstehoek (Fig. 7c), Steynsdorp and Doornhoek (Fig. 7e) and some samples from
Kaap Valley exhibit flatter trends than the rest of the
plutons.
Depletion in the HREE, with respect to chondritic
concentrations, is usually interpreted as a source-related
feature due to preferential partitioning of these elements
into coexisting restitic garnet. Rapp et al. (1991) produced TTG-like liquids with YbN < 3 in his high-T, fluidabsent partial melting experiments on metabasalts, at
pressures within the garnet stability field. Although YbN
varies from about 1–10 in individual Barberton TTG
samples, the plutons have average YbN values varying
from 1.19 (Theeboom) to 3.84 (Kaap Valley), with an
overall average of 2.61. The average Archaean amphibolite (Gao et al., 1998) has YbN ≈ 12, which suggests
that the Barberton TTG magmas were mostly depleted in
HREE with respect to their possible source rocks, though
by varying amounts for different plutons. This is consistent with garnet-bearing residues, with some variation in
the proportion of garnet.
Published REE partition coefficients for hornblende
and plagioclase (summarised in Rollinson, 1993) suggest that the presence of a large amount of hornblende as
a residual mineral in the magma source could negate the
influence of a small amount of residual plagioclase, producing a melt lacking a negative Eu anomaly. However,
Rapp and Watson (1995) showed that fluid-absent partial melting of metabasic rocks only produces TTG-like
liquids (and melts of any significant quantity) at temperatures above amphibole stability. Thus, we infer that
large amounts of residual hornblende are unlikely to have
been present. Thus, the observed lack of Eu anomalies is
Pluton
9.933
12.089
7.683
7.240
8.890
7.208
7.810
8.980
Nd (ppm)
1.518
2.120
1.884
1.210
1.520
1.387
1.550
1.610
Sm (ppm)
0.09236
0.10596
0.14817
0.10070
0.10330
0.11631
0.12020
0.10840
147 Sm/144 Nd
rock(0)
0.510323
0.510618
0.511474
0.510448
0.510566
0.510740
0.511076
0.510814
143 Nd/144 Nd
rock(0)
20
15
41
35
23
27
24
10
±2 se
0.508348
0.508352
0.508305
0.508153
0.508211
0.508090
0.508339
0.508345
143 Nd/144 Nd
rock(t)
3.236
3.236
3.236
3.445
3.445
3.445
3.443
3.443
t (Ga)a
−1.64
−1.56
−2.48
−0.07
1.09
−1.30
3.54
3.67
␧Nd(t)b
63
Sample
Nelshoogte
Nelshoogte
Nelshoogte
Stolzburg
Stolzburg
Theeboom
Theespruit
Theespruit
Table 1
Nd isotope data for Barberton TTG plutons
NLG1
NLG13
NLG25a
STZ1
STZ18
TBM1b
TP21c
TP30c
Systematics of the Sm–Nd isotope system are used
to calculate ␧Nd at the known crystallisation ages of the
plutons concerned (Kamo and Davis, 1994). The calculated ␧Nd(t) values are presented in Table 1 and Fig. 7,
and range from +3.67 to −2.48. For comparison, the
␧Nd(t) values for the greenstone volcanic samples from
the Onverwacht Group (Hamilton et al., 1979) are also
included. These were calculated using recent U-Pb zircon dates (Armstrong et al., 1990). Also marked on the
diagram are lines representing CHUR (␧Nd = 0) and the
evolution of the depleted mantle, as defined by Goldstein
et al. (1984), which assumes a linear increase in ␧Nd,
from a chondritic value at 4.5 Ga to a present day value
of +10.
From Fig. 8 it is clear that most of the analysed
∼3.45 Ga TTGs have positive ␧Nd values. This suggests that they were derived either from the mantle
directly, or from a juvenile crust that had not long been
separated from the mantle. This is consistent with the
published view that the Theespruit Formation represents
an accreted oceanic arc fragment (De Wit et al., 1987).
The Stolzburg pluton, in particular, exhibits a close similarity to the basalts and komatiites of the Theespruit
(Tspt), Komati (Kom) and Hoogenoeg (Hoog) Formations (Hamilton et al., 1979). Its values typically lie
between those of CHUR and the depleted mantle. On this
evidence, it seems possible that greenstone rock types
may be the sources of the TTG magmas.
3.6. Nd isotopes
cited above suggest that this melting occurred in thickened crust. The Steynsdorp and Doornhoek magmas may
have been derived from slightly shallower depths, where
residual garnet was present in smaller quantities, perhaps
accompanied by a small amount of plagioclase.
a U-Pb zircon ages from Armstrong et al. (1990), Kamo and Davis (1994), Kröner et al. (1991, 1996).
b Calculated using present day chondritic values of 143 Nd/144 Nd = 0.512638 and 147 Sm/144 Nd = 0.1967; 147 Sm/144 Nd
rock(0) = the Sm/Nd isotopic
ratio of the rock sample at present time; 143 Nd/144 Ndrock(0) = the Nd isotopic ratio of the rock sample at present time (i.e. t = 0); 143 Nd/144 Ndrock(t) = the
Nd isotopic ratio of the rock sample at time t; ␧Nd(t) = [143 Nd/144 Ndrock(t) /143 Nd/144 NdCHUR(t) − 1] × 104 .
c Samples collected by C. Anhaeusser.
source rocks with production of a garnet-bearing restite.
In general, plagioclase was not a major phase in the
residual source nor was it a fractionating phase during
magma evolution. The exceptions are the Steynsdorp and
Doornhoek plutons, in which there is some geochemical
evidence for minor plagioclase fractionation or perhaps
a small amount of plagioclase in the residuum. The relatively flatter REE patterns of rocks from these two
plutons imply that the restite produced during their generation was relatively garnet-poor. There is geochemical
evidence for the presence of a minor metasedimentary
component in the TTG magma source region, specifically for the rocks of the Nelshooghte pluton. The compositions of the residua are important because their mineralogy provides constraints on the depths (pressures)
at which the TTG magmas were generated. Garnet is
stable at relatively high pressures in metabasic rocks,
such as the sources of TTG magmas. For fluid-absent
partial melting of metabasites, at 1000–1150 ◦ C, Rapp
et al. (1991) and Rapp and Watson (1995) showed that
pressures >0.8 GPa (about 30 km) are required to stabilise garnet, and ≥1.2 GPa (about 40 km) for garnet
to be stable in the absence of plagioclase. Plagioclase
disappears from the phase assemblage by dissolution
in the melt, at high T, and by breakdown to jadeitic
pyroxene and the grossular component in garnet, towards
high P. The Rapp and Watson (1995) experiments on
an Archaean greenstone (Fig. 2, op. cit.) show a window for simultaneous garnet presence and plagioclase
absence at T > 1000 ◦ C and P between 1.2 and 2 GPa. As
noted above, the chemistry of most of the Barberton TTG
rocks strongly suggests that the melts equilibrated with
plagioclase-free, garnet-bearing residual source assemblages. From this we suggest that most of the Barberton
TTG rocks were derived from sources located at depths
of at least 40 km. The metamorphic and structural studies
64
were present at the required depths. REE data suggest
that the TTG magmas were mostly generated at depths
where garnet was stable and formed part of the melting
residue.
Trace- and rare earth-element modelling is helpful
in constraining the actual composition of the restite.
Fig. 9 compares the chondrite-normalised La/Yb ratios
((La/Yb)N ) and Sr/Y ratios of the Barberton granitoids.
It highlights the fact that the TTG suite has low YbN and
Y values and highly evolved (La/Yb)N and Sr/Y ratios,
typical of Archaean TTGs (Martin, 1986). The arrows on
the diagrams represent batch melting trends for Archaean
tholeiites that produce restite compositions of amphibolite, and 7–30% garnet-rich amphibolite or eclogite,
as modelled by Petford and Atherton (1996), Atherton
and Petford (1993), Martin (1986) and Drummond and
Defant (1990). Due to a lack of data for Barberton amphibolites, we cannot calculate a model that is more specific
to this area. However, the differences are unlikely to
invalidate the general observations. Using the literaturederived trends as a guide, it appears that the TTG suite
could have been derived through partial melting of a
primitive basaltic source, producing an amphibolitic or
eclogitic restite with >30% garnet (Fig. 9).
The likelihood that the Barberton TTGs represent
relatively little-modified initial magma compositions,
coupled with the inference of an initial magma in equilibrium with garnet, raises the possibility of establishing the
minimum pressure of magma genesis, using experiments
to establish the lowest pressure at which garnet would
be stable in the TTGs. This is perhaps a more accurate
way of determining pressure of genesis than establishing
the pressure of garnet appearance in a general metamafic
protolith. This is because the specific protolith composition is unknown, and garnet stability is sensitive to a
Fig. 8. Graph showing ␧Nd vs. age (Ga) of the Barberton granitoid
rocks. BK and BKT: members of the potassic granitic suite, NLG:
Nelshoogte, STZ: Stolzburg, TBM: Theeboom, TP: Theespruit, Tspt:
Theespruit Formation, Kom: Komati Formation, Hoog: Hoogenoeg
Formation.
3.7. Implications of the geochemical results
The lack of major- and trace-element evidence for
mineral fractionation in the petrogenesis of most of the
Barberton TTG plutons suggests that the most of the
measured rock compositions probably reflect magma
compositions. This is consistent with the interpretation
that the differences in chemistry between plutons of
the same age also reflect separate, contrasting magma
batches, and that the scattered trends within the plutons probably reflect heterogeneities within their source
rocks, retained through incomplete magma homogenisation. The ␧Nd values suggest that the TTGs were
derived from juvenile crustal sources with depletedmantle signatures. The similarity between the ␧Nd values of the ∼3445 Ma TTGs and the Lower Onverwacht
greenstones further implies that the greenstone materials could be the sources of these magmas, if they
Fig. 9. (a) (La/Yb)N –YbN and (b) Sr/Y–Y diagrams for the TTG rocks. Arrows indicate the batch melting pathways of Archaean tholeiites with
source mineralogies of amphibolite, 7–30% garnet-rich amphibolite or eclogite (adapted from Petford and Atherton, 1996). Archaean TTG/high-Al
trondhjemite-tonalite-dacite (TTD) and post Archaean granite/andesite-dacite-rhyolite (ADR) fields adapted from, e.g. Martin (1986), Drummond
and Defant (1990), Atherton and Petford (1993), Petford and Atherton (1996). Dark shading: overlap area between the TTG and ADR fields.
65
q
or
ab
an
di
hy
il
ap
SiO2
TiO2
Al2 O3
FeOT
MnO
MgO
CaO
Na2 O
K2 O
P2 O5
49
0.27
0.98
22.01
9.24
48.54
13.34
0.27
5.94
0.47
0.18
70.33
0.24
16.04
2.00
0.03
1.11
2.85
5.74
1.56
0.08
THE4A
66
0.27
0.65
1.89
3.25
17.97
25.96
15.48
33.56
1.50
0.39
53.81
0.79
13.71
9.13
0.28
10.09
9.41
2.05
0.55
0.17
AmX12-a
54b
0.32
0.66
6.14
3.19
14.76
25.38
14.27
33.70
2.17
0.39
54.31
1.14
12.85
12.13
0.29
7.97
8.92
1.68
0.54
0.17
Ave greenstonea
4.1.2. Partial melting of amphibolite
Partial melting of amphibolitic rocks has been well
studied since the 1970s, with an emphasis on fluid-absent
phase relations since the 1990s. Recently, Clemens
(2005) and Moyen and Stevens (2005) summarised the
data on phase relations, melt proportions and their evolution with T, and melt compositions. For fluid-absent
conditions, Moyen and Stevens (2005) found that, at T
near 900 ◦ C, the pressure of the garnet-in phase boundary varies between <1.0 and 1.4 GPa, as a function of the
composition of the starting rock. Likewise, the positions
of the solidus and the amphibole-out boundaries also
vary substantially. Thus, despite the volume of this previous work, it seems useful to experimentally investigate
the partial melting behaviour of the Barberton amphibolite. This is because these rocks are relatively low in
Al2 O3 , and have high Mg#s, features uncommon among
previously studied compositions.
The Theespruit Formation is the portion of the Barberton greenstone belt that experienced the highest
grade of metamorphism, and is the only part of the
and mineral geochemical compositions are shown in
Tables 2 and 3, respectively.
FeOT = total Fe as FeO.
a Average of nine analyses.
b Range = 20–71.
Mg#
K2 O/Na2 O
A/CNK
Table 2
Major element and CIPW normative compositions (100% anhydrous)
of the starting materials for the experiments and the average Theespruit
Formation greenstone
number of subtle geochemical variations in metabasic
rocks.
4. Experiments
4.1. Starting materials
4.1.1. Garnet stability in TTG rocks
A specimen from the Theespruit Pluton (THE4A) was
chosen as the starting material for this series of experiments. We selected a rock that we judged most likely to
be representative of a TTG magma that had undergone
little modification by fractional crystallisation. Thus, the
sample has relatively low SiO2 (∼70 wt%), high Mg#
(49), normal K2 O (∼1.6 wt%) and P2 O5 (0.08 wt%),
and minimal evidence for secondary alteration. THE4A
contains plagioclase (An15 ), quartz, biotite (Mg# = 52),
minor hornblende (Mg# = 54), minor microcline (Or96 )
and accessory apatite, allanite and epidote (Fig. 10).
Although the rock shows signs of alteration, with minor
sericitisation of plagioclase, it represents the freshest
sample collected from the Theespruit pluton. Bulk rock
Fig. 10. Photographs illustrating the mineralogy of experimental starting material THE4A, (a) in plane polarised light and (b) under crossed
polars. Pl: plagioclase, Mc: microcline, Qtz: quartz, Bt: biotite, Hbl:
hornblende, Ap: apatite, Ep: epidote, Al: allanite.
66
Table 3
Compositions of minerals in the starting materials
Hbl
4
36.93
1.66
15.58
0.10
0.00
19.64
0.33
11.92
0.20
0.21
9.43
96.00
Bt
8
2.862
0.003
1.130
–
–
0.003
–
0.007
0.151
0.827
0.007
4.99
–
An15
2
64.95
0.10
21.76
–
–
0.08
–
0.11
3.20
9.68
0.11
100.01
Pl
8
2.983
0.001
1.003
–
0.005
–
–
–
0.002
0.040
0.975
5.02
–
Or96
2
64.11
0.02
18.29
–
0.15
–
–
0.00
0.05
0.45
16.42
100.21
Kfs
23
6.679
0.104
1.779
0.015
0.356
1.249
0.035
2.782
1.908
0.352
0.114
15.37
68
–
6
46.74
0.97
10.56
0.14
3.30
10.44
0.29
13.06
12.46
1.27
0.63
99.86
Hbl
8
2.924
0.006
1.077
–
–
0.008
–
0.002
0.058
0.898
0.017
4.99
–
An6
2
67.94
0.19
21.24
–
–
0.21
–
0.03
1.27
10.76
0.32
101.94
Pl
AmX12-a
1
46.13
0.24
9.51
0.08
4.26
15.27
0.45
10.18
12.40
1.19
0.88
100.59
22
5.417
0.183
2.695
0.011
–
2.411
0.042
2.608
0.032
0.061
1.762
15.24
52
THE4A
n
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
Total
23
6.730
0.026
1.635
0.009
0.467
1.863
0.056
2.214
1.938
0.337
0.164
15.44
54
with the average of nine other mafic rocks from the
Theespruit Formation. Its Mg# lies close to the middle
of the wide range for these rocks (footnote in Table 2)
and, apart from marginally higher Na2 O and lower
TiO2 , AmX12-a is typical of Theespruit mafic rocks,
and indeed of some other Barberton greenstones. Partial
melting experiments on this material, at appropriate P, T
and fluid conditions should therefore provide us with an
answer to the question of whether these rocks could have
been the source for the Barberton TTGs. They also make
a useful contribution to the overall dataset on amphibolite melting relations.
Most amphibolite-facies Barberton greenstones
carry some greenschist-facies retrograde overprint. In
AmX12-a this was marked by minor replacement of
hornblende by chlorite and plagioclase by epidote. Since
the presence of these lower-grade minerals increases
the H2 O content of the sample, over that which would
apply to progressively metamorphosed equivalents in
the upper amphibolite facies, the sample needed to be
xO2−
Si
Ti
Al
Cr
Fe3+
Fe2+
Mn
Mg
Ca
Na
K
Cations
Mg#
Fsp comp.
Onverwacht Group whose mineral assemblages record
upper amphibolite-facies metamorphic conditions. This
results from the fact that it formed part of the ‘lower
plate’ during 3.23 Ga tectonism, was decoupled from the
rest of the greenstone belt and experienced a separate,
higher-grade history, along with the high-grade southern granitoid gneiss terrain (Dziggel et al., 2002; Kisters
et al., 2003; Diener et al., 2005). Felsic volcanic rocks
within the Theespruit Formation show strong geochemical similarities with the TTG rocks of the Steynsdorp
pluton (e.g. HREE depletion and absence of any marked
Eu anomaly; Diener, 2004), and so may be older that the
Stolzburg and Theespruit plutons. These relationships
indicate that mafic rocks from the Theespruit Formation
could be the sources for both the 3.45 and 3.23 Ga TTG
magmas.
An amphibolite sample (AmX12-a) was therefore
taken from one of the large greenstone remnants enclosed
within the Batavia pluton (Fig. 1). For purposes of comparison, its bulk composition is given in Table 2, along
67
4.2.2. Partial melting of amphibolite
We also took the Barberton basaltic greenstone
amphibolite (AmX12-a) and subjected it to the metamorphic conditions suggested by the results of the TTG
near-liquidus experiments. The compositions of partial
melts of this basaltic amphibolite can be compared with
the compositions of the TTG rocks, to determine whether
Barberton amphibolites, such as this, could have been
the sources of some of the Barberton TTG magmas. The
pressure used for the partial melting experiments was
based on the results of the near-liquidus experiments.
Conditions were set at 1.6 GPa, and melting temperatures of 875, 925 and 1000 ◦ C were investigated. This
temperature range is appropriate to the production of
partial melts from amphibolitic source materials (see,
e.g. Clemens, 2005; Moyen and Stevens, 2005), and
allowed us to assess the effect of T on the degree of
melting, the compositions of melts, and stability ranges
of minerals in the melting interval. In fluid-absent partial melting, the hydrous character of any partial melt
is derived through breakdown of crystalline hydrates,
in this case hornblende. Thus, no additional H2 O was
added to the experimental charges in this series of
experiments.
(at 900 ◦ C and 1 GPa). The H2 O content of the bulk
rock derived from biotite (∼15 vol%) already present in
the starting material, would be about 0.6 wt%. Thus, we
would need to add 5 wt% H2 O to an experimental charge,
in order to obtain approximately the correct overall H2 O
content. We chose to use an experimental charge of 0.01 g
of TTG rock powder so this would require the addition
of ∼0.5 ␮L (i.e. 5 wt%) of high-purity deionised H2 O.
The first experiment, using 0.5 ␮L of added H2 O, and
run at 875 ◦ C and 1 GPa, produced only glass, indicating
super-liquidus conditions. The H2 O content was subsequently reduced to 0.4 ␮L (i.e. 4 wt%), which ensured
that there would be at least some crystals present. Note
that no free fluid phase was present in any of the nearliquidus TTG experiments; all H2 O was dissolved in the
melt phase.
The philosophy behind these phase equilibrium
experiments is that, at the pressure and temperature at
which a magma is generated, the melt will be in equilibrium with the mineral phases that constitute the restite.
If the magma then remained near its site of generation
and crystallised, the near-liquidus phases would mimic
the phases in the restite. Thus, if we can determine the
minimum pressure at which garnet appears near the liquidus of a Barberton TTG, we will obtain an estimate
of the minimum pressure at which the TTG magma was
generated.
amphibolitised prior to using it as a starting material.
This was achieved by placing it in a gold capsule, and
annealing it in an internally heated gas vessel, at 650 ◦ C
and 0.2 GPa, for 7 days. The resulting mineralogy was
∼70% amphibole (Mg# = 68) and ∼30% plagioclase
(An6 ), with traces of K-feldspar. Bulk-rock and mineral chemical compositions are given in Tables 2 and 3,
respectively.
4.2. Pressure, temperature and ﬂuid conditions
4.2.1. Garnet stability in TTG
In metaluminous rocks, the stability of Mn-poor garnet depends primarily on pressure. Therefore, we kept
the temperature constant in this first series of experiments. A temperature of 875 ◦ C was chosen because this
would be near the TTG liquidus at the chosen H2 O content of the system (see below). Pressure was initially
set at 1 GPa and then varied, by increments of 0.1 GPa,
between 1.2 and 1.7 GPa, until the initial appearance of
garnet was tightly constrained.
A key assumption of this work is that a source rock of
amphibolitic composition partially melted, under fluidabsent conditions, to form TTG magma. In the Archaean,
with the possibility of higher geothermal gradients than
at present, this sort of high-T melting could conceivably occur in a subducting slab (e.g. Martin, 1994), in
the roots of oceanic plateaux (e.g. Condie, 2005) or in
the deep crust of the thickened upper plate (e.g. Petford
and Atherton, 1996). Stevens and Clemens (1993) and
Clemens and Watkins (2001) reviewed the evidence for
the dominance of fluid-absent conditions during the formation of granitoid magmas by crustal melting. Under
these conditions, the hydrous minerals provide the H2 O
component required to form hydrous granitoid melts.
The magma from which THE4A crystallised would
have contained far more H2 O than is now present in
the hydrous minerals of the rock. The average initial
H2 O content of granitoid magmas is ∼4 wt% (Clemens,
1984). Thus, H2 O must be added to the system to replicate the true initial TTG magma composition.
An estimate of the amount of H2 O in the assumed
amphibolitic source rock leads to an approximation of
the amount of H2 O present in the melt. Calculations
were made on the basis of 25% partial melting of amphibolite, at 900 ◦ C and 1 GPa, with complete destruction
of the hornblende. Using the model of Clemens and
Vielzeuf (1987), this would require 1.4 wt% of H2 O in
the source rock. For fluid-absent melting, melt proportion (wt%) = H2 O content of rock/H2 O content of melt
formed (Clemens and Vielzeuf, 1987). Thus, the amount
of H2 O required in the TTG melt would be 5.6 wt%
68
4.3. Experimental methods
±
±
±
±
±
±
±
±
±
1
1
1
1
1
1
1
1
1
96
96
96
96
96
112
96
96
96
Duration (h)
melt
Cpx, Hbl, melt
Hbl, melt
Cpx, Hbl, melt
(Grt), Hbl, melt
Grt, Cpx, Hbl, Ru, melt
Grt, (Ru), (Ap), Opx, Cpx, Hbl, Qtz, melt
Grt, (Ru), (Ap), Opx, Cpx, Hbl, melt
Grt, (Ru), (Ap), Opx, Cpx, melt
Run products
All experimental conditions and run products are
given in Table 4. Note that because the starting material was crystalline (effectively seeded with plagioclase), we can be reasonably confident that the absence
of plagioclase in the run products reflects its instability under these P–T–aH2 O conditions. The pressure for the appearance of near-liquidus garnet is constrained to 1.52 ± 0.05 GPa, which equates to a depth of
51 ± 2 km (assuming a largely mafic crust with a density
of 3000 kg/m3 ). Due to the importance of the minimum
pressure for the appearance of near-liquidus garnet, we
compared the experimental result with the prediction
of the Perple X linear programming package (see, e.g.
Connolly, 2005) for calculation of phase diagrams using
thermodynamic properties of minerals and melts. We
used the bulk composition of THE4A, with total H2 O in
the system set at 4.42 wt%, as in the experiments. Phases
considered in the calculations were Qtz, Pl, Kfs, Bt, Hbl,
Grt, Cpx, Opx and melt. The silicate melt composition
was fixed as the composition found experimentally, in
run PC1-073 (Table 5). The predicted pressure for the
5.1. TTG near-liquidus experiments
5. Results
were run for 96 h and quenched isobarically, until T fell
to about 600 ◦ C, to prevent vesiculation of any glass
present, as a result of volatile exsolution. When the runs
were completed, the capsules were removed from the
cells, cleaned, opened and the contents examined. A part
of each run product was hand-ground in an agate mortar
and a grain mount made, for examination with an optical microscope. Another intact piece of the run product
was mounted in epoxy resin, sectioned and polished for
analysis with the SEM/microprobe.
P (MPa)
875
875
875
875
875
875
875
925
1000
T (◦ C)
The powdered starting materials were finely ground,
in a mechanical agate mortar, to an average grain size
of ∼5 ␮m, dried at 110 ◦ C and stored, over silica gel, in
a vacuum desiccator. Gold capsules (10 mm long, 3 mm
OD and 0.15 mm wall thickness) were used to contain the
samples plus any added H2 O. Au is unreactive, does not
strongly absorb Fe from the sample, inhibits H2 diffusion
and thus preserves fO2 at realistic reducing levels. Based
on previous studies of the redox conditions in this pistoncylinder apparatus, log fO2 is believed to lie between
QFM and QFM-2 (Graphchikov et al., 1999). Each capsule was annealed and arc welded at one end prior to
the addition of ∼0.01 g of powdered sample. For experiments with added H2 O, deionised water was loaded into
the capsule, before the powder, using a 10 ␮L syringe.
These capsules were sealed by arc welding with the lower
end submerged in a water bath, to cool the capsule and
prevent boiling and loss of fluid. For the fluid-absent
melting experiments on the greenstone amphibolite, the
powder was added, the open end of the capsule gently
crimped shut and the capsule dried, at 110◦ for 30 min,
prior to final arc welding.
Experiments were carried out in a 12.7 mm diameter, non-end-loaded, piston-cylinder apparatus (Depths
of the Earth Company, Tempe, Arizona, USA). NaClPyrex cells were used for experiments above 900 ◦ C. For
lower-temperature runs NaCl-only cells were used. Thermocouples were type-K (chromel-alumel), and temperatures are considered to be accurate to ±1 ◦ C. Precision
in pressure control varied between different experiments
but was generally around ±0.05 GPa. The Au capsules
were flattened and folded into small packets, the sides
of which rested flat in the cells, separated from the thermocouple tip by a thin disk of alumina. Experiments
Sample composition
Table 4
Experimental conditions and run products
Run no.
989
1184
1407
1474
1574
1717
1598
1596
1581
23
12
22
22
22
29
24
20
47
THE4Aa + 0.5 ␮L H2 Ob
AmX12-a
AmX12-a
AmX12-a
±
±
±
±
±
±
±
±
±
PC1-069
PC1-070
PC1-075
PC1-077
PC1-080
PC1-073
PC1-083
PC1-086
PC1-087
NB parentheses indicate minor amounts of the phase present.
0.01 g powdered rock.
H2 O added to capsules but all runs fluid-absent at P and T (with ∼4 wt% or ∼5 wt% H2 O in the melt, near the liquidus).
a
b
PC1-069
1.2
6
71.53
0.30
16.25
1.26
0.06
0.66
2.32
5.92
1.71
48
0.29
1.03
PC1-070
1.4
6
71.07
0.34
16.01
1.64
0.03
0.90
2.78
5.61
1.63
49
0.29
1.00
PC1-075
1.5
6
71.91
0.28
16.27
1.49
0.06
0.66
2.51
5.17
1.67
44
0.32
1.09
PC1-077
43.78
5.26
9.28
25.98
11.81
1.6
5
74.37
0.19
16.64
1.27
0.09
0.42
2.38
3.07
1.57
37
0.51
1.50
PC1-080
2.86
0.51
31.26
2.19
9.75
42.05
11.36
1.7
8
72.86
0.27
16.32
1.10
0.09
0.45
2.29
4.97
1.65
42
0.33
1.16
PC1-073
69
23.78
27.99
1.39
9.87
43.75
12.45
3.23
0.36
9.63
47.47
13.69
0.08
4.71
0.65
Fig. 12. Average glass (quenched melt) compositions produced in
near-liquidus experiments on THE4A, at a range of pressures, plotted on the classification diagram of Barker (1979).
that the TTG magmas were generated in a geothermal gradient <20 ◦ C/km. Coexisting minerals are listed
in Table 6. Typical near-liquidus ferromagnesian mineral assemblages were Hbl ± Cpx at P ≤ 1.5 GPa and
Hbl + Cpx + Grt at P ≥ 1.6 GPa (all mineral abbreviations from Kretz, 1983). Plagioclase was not present in
4.03
0.53
Table 5
Average major-elementa and CIPW normative compositions of the melts produced in near-liquidus experiments on THE4A
1.0
4
71.48
0.36
16.38
1.99
0.04
1.08
2.84
4.21
1.62
49
0.38
1.18
23.71
0.44
10.11
50.09
11.51
Run no.
P (GPa)
n
SiO2
TiO2
Al2 O3
FeOT
MnO
MgO
CaO
Na2 O
K2 O
Mg#
K2 O/Na2 O
A/CNK
31.67
2.54
9.57
36.19
13.52
3.57
0.57
Normalised to 100% anhydrous, n = no. of analyses.
5.82
0.68
a
q
c
or
ab
an
di
hy
il
incoming of garnet, at 875 ◦ C, is 1.45 GPa—very close
to the pressure found experimentally.
Reaction products and their relative proportions vary
with P, as shown in Fig. 11. The amount of melt produced was ≥75% for all runs and, as expected, melt
compositions (Table 5) are generally trondhjemitic to
granodioritic (Fig. 12). The high percentage of melt at
875 ◦ C suggests that the TTG magmas were probably
produced from their protoliths at temperatures around
900 ◦ C, and very probably less than 1000 ◦ C. Taking
the inferred depth of origin into account, this implies
Fig. 11. Graph showing the proportions of phases in the products of
near-liquidus experiments on THE4A. See text for further explanation.
70
1.4
10
50.24
0.80
8.12
0.17
6.20
4.91
0.18
16.08
10.59
2.32
0.39
1.5
16
49.10
1.23
10.20
0.08
4.29
6.90
0.12
14.53
10.54
2.64
0.38
23
6.399
0.153
2.017
0.011
0.917
0.575
0.016
2.688
1.385
0.607
0.189
14.97
82
1.6
17
46.42
1.48
12.40
0.11
8.90
4.93
0.14
13.09
9.36
2.29
0.82
23
6.858
0.153
2.406
0.007
0.218
1.075
0.011
2.268
1.327
0.917
0.119
15.36
68
1.7
17
48.94
1.45
14.56
0.06
2.05
9.15
0.09
10.83
8.81
3.38
0.66
12
3.032
0.056
1.911
0.007
0.067
1.380
0.065
0.719
0.799
0.045
0.008
8.089
34
1.7
16
38.94
0.95
20.84
0.11
1.16
21.20
0.98
6.20
9.58
0.29
0.08
Grt
6
1.899
0.009
0.102
0.003
0.059
0.148
0.005
0.744
0.858
0.073
0.003
3.91
83
1.2
7
52.64
0.32
2.40
0.12
2.16
4.94
0.16
13.84
22.20
1.04
0.07
Cpx
6
1.973
0.004
0.085
0.003
–
0.416
0.007
0.811
0.491
0.055
0.000
3.85
66
1.5
1
54.95
0.16
2.01
0.12
–
13.84
0.23
15.15
12.77
0.79
0.01
6
1.950
0.006
0.218
–
–
0.473
0.012
0.747
0.502
0.082
0.020
4.01
61
1.7
1
52.00
0.20
4.93
–
–
15.09
0.39
13.36
12.48
1.13
0.41
Amp
1.2
4
49.47
1.23
8.81
0.09
6.04
5.23
0.13
15.68
10.41
2.55
0.35
23
6.751
0.127
1.652
0.009
0.441
0.795
0.014
2.977
1.552
0.704
0.066
15.09
79
Table 6
Average major-element compositionsa of minerals produced in near-liquidus experiments on THE4A
P (GPa)
n
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
23
6.954
0.083
1.323
0.019
0.642
0.574
0.021
3.316
1.572
0.622
0.070
15.20
85
In view of the results of the TTG near-liquidus experiments, the greenstone amphibolite experiments were all
conducted at 1.6 GPa; results are shown in Table 4. Note
5.2. Partial melting of Barberton amphibolite
with the original edenitic hornblende in the starting material (THE4A). Garnet formed at 1.7 GPa has a composition of Alm48 Prp24 Grs26 Sps2 . Garnet formed at lower
pressures was scarce, but easily identified in optical grain
mounts, though it was not found during microprobe analysis. Pyroxenes are typically diopsidic at P = 1.2 GPa and
augitic at P ≥ 1.5 GPa. A very small amount of accessory
rutile was identified in the products of the run at 1.7 GPa
(Fig. 13), and may have been present in the other experiments, though it was not specifically identified. Residual
rutile is required to explain the marked negative Nb,
Ta and Ti anomalies shown by all TTG rocks (see, e.g.
Fig. 5). The lack of plagioclase and presence of garnet
and pyroxene in the near-liquidus assemblage suggest
that the majority of TTGs coexisted with a hornblendebearing eclogite rather than either an amphibolitic or
granulitic residue.
Normalised 100% anhydrous, Fe2+ and Fe3+ calculated using the method of Droop (1987).
23
6.817
0.128
1.431
0.010
0.626
0.604
0.016
3.220
1.536
0.681
0.062
15.13
84
a
xO2−
Si
Ti
Al
Cr
Fe3+
Fe2+
Mn
Mg
Ca
Na
K
Cations
Mg#
the near-liquidus assemblages, confirming the interpretation based on the lack of a Eu anomaly, that feldspar
was not present in the residuum. Textures in the run products are exemplified in the photomicrograph of the run
product formed at 1.7 GPa (Fig. 13). Mineral compositions vary little with P. The amphiboles range from silicic
edenite to magnesiohastingsitic hornblende, compared
Fig. 13. Back-scattered electron photomicrograph of run PC1-073
(1.7 GPa) showing new garnet crystals in matrix glass (quenched melt).
71
with T, from ∼5% at 875 ◦ C to 30% at 1000 ◦ C. Melt
compositions are given in Table 7.
The compositions of coexisting minerals are
given in Table 8. Full mineral assemblages are;
Amp + Cpx + Opx + Grt + Qtz + Rt + Ap at 875 ◦ C,
Amp + Cpx + Opx + Grt + Rt + Ap at 925 ◦ C and
Cpx + Opx + Grt + Rt + Ap at 1000 ◦ C (Fig. 14). The
amphiboles are edenites to edenitic hornblendes, similar
to those of the starting material. In the 875 ◦ C run,
large original hornblende crystals remain (Fig. 14a),
indicating incomplete breakdown, due to the low
degree of partial melting. Compositions of garnets are
typically Alm-rich with Prp content increasing at higher
T (Alm46 Prp29 Grs22 Sps3 to Alm38 Prp38 Grs23 Sps1 )
(Fig. 14b). Pyroxenes are typically augite, coexisting
with enstatite, at all temperatures. Previous studies of
fluid-absent partial melting of amphibolites, at about
1.5 GPa, produced pyroxenes with Jd contents, ranging
from 8 to 14 mol.% with an average of 10 mol.% (Sen
and Dunn, 1994; Rapp and Watson, 1995; Skjerlie and
Patiño Douce, 2002). The Jd contents of the clinopyroxenes formed in the present work are 6–7.5 mol.%,
similar to those formed in the near-liquidus runs on
the TTG starting material. This difference is probably
related to the bulk composition of the starting material
and the particular phases coexisting with the clinopyroxenes. Experiments on a different amphibolitic starting
material, carried out in the same apparatus, with the
same experimental techniques, at 2 and 2.5 GPa (Xiao
and Clemens, submitted for publication), produced
strongly omphacitic pyroxenes coexisting with garnet
and melt. In the present work, sodic hornblende was
present, and at least as abundant as the coexisting
clinopyroxene, up to about 920 ◦ C (Fig. 15). Partitioning of Na between the amphibole and the pyroxene is
probably a major control on the lower Jd contents of the
Fig. 15. Proportions of run products formed in the partial melting
experiments on greenstone amphibolite AmX12-a.
Fig. 14. Back-scattered electron photomicrographs showing the mineralogy and textures of the products of partial melting experiments
(1.6 GPa) carried out on greenstone amphibolite AmX12-a, (a) at
875 ◦ C, (b) at 925 ◦ C and (c) at 1000 ◦ C.
that we have not attempted to determine the minimum
pressure for appearance of garnet in this amphibolite, as
this is not relevant to the problem being investigated. The
pressure for the partial melting experiments was simply
chosen as a suitable pressure, slightly above the necessary minimum, as determined from the near-liquidus
experiments on the TTG (THE4A). Fig. 14 shows typical
textures developed in the run products. Relative proportions of the phases in the run products are shown, as a
function of T, in Fig. 15. The amount of melt increases
72
X̄
σ
X̄
0.27
0.08
0.10
0.18
0.08
0.15
0.18
0.13
σ
T = 925 ◦ C, n = 11
74.57
0.33
14.05
1.59
0.27
1.83
4.24
3.10
X̄
0.30
0.06
0.09
0.19
0.16
0.12
0.05
0.13
σ
T = 1000 ◦ C, n = 5
70.03
0.69
15.35
2.69
0.73
2.84
4.56
3.12
T = 1000 ◦ C, n = 5
0.68
0.96
T = 925 ◦ C, n = 11
23.06
18.44
39.13
11.66
–
1.60
4.81
1.31
32.6
23.0
0.73
1.02
T = 875 ◦ C, n = 7
32.65
18.32
36.30
8.66
.39
–
3.05
0.63
23.2
19.3
0.34
0.09
0.20
0.17
0.10
0.13
0.17
0.12
0.66
0.91
35.64
17.14
37.44
7.44
–
3.12
1.60
0.61
26.3
16.6
74.94
0.32
13.21
1.60
0.32
2.31
4.38
2.90
T = 875 ◦ C, n = 7
Table 7
Normalised 100% anhydrous analyses and CIPW norms of glasses (quenched melts) formed in fluid-absent partial melting experiments (1.6 GPa)
on amphibolitised Barberton greenstone AmX12-a
SiO2
TiO2
Al2 O3
FeOT
MgO
CaO
Na2 O
K2 O
K2 O/Na2 O
A/CNK
q
or
ab
an
c
di
hy
il
Mg#
An in plag.
Variations in melt composition with run T reflect
the changes in stoichiometry of the incongruent melting reaction, as melting progresses. For example, SiO2
contents of melts decrease, whereas the other oxides
generally increase with increasing T. The reduction in
SiO2 and increase in CaO, FeOT and TiO2 reflect the
increased degree of melting/dissolution of mafic phases
as T increases. However, there is no progressive increase
in CaO between 875 and 925 ◦ C. This probably reflects
the stability of clinopyroxene at these temperatures.
Once T exceeds 925 ◦ C, Cpx becomes unstable and then
contributes Ca to the melt. Increases in Na2 O, with temperature, are typically ascribed to the melting of Na-rich
minerals, the identity of which will change as a function of P and T. Plagioclase is not present in the residual
assemblages at the conditions of our experiments, having been consumed in the above melting reaction by
875 ◦ C. Increases in melt Na2 O content at higher T were
therefore most probably the consequence of amphibole
dissolution. Similarly, the trace of K-feldspar present in
the starting material had disappeared by 875 ◦ C. Despite
this, higher temperature evolution to significantly larger
melt volumes does not markedly shift the melt composition to less potassic (lower K2 O/Na2 O) compositions
FeOT = total Fe as FeO; n = number of individual point analyses; X̄ = sample mean; σ = standard deviation.
clinopyroxenes in the present work. In the products of
the lower-T melting experiments, the two pyroxenes are
quite complexly intergrown (see, e.g. Fig. 14b) while, at
1000 ◦ C, the pyroxene crystals are more separated from
each other (e.g. Fig. 14c). At 1.6 GPa, the quartz-out
curve lies between 875 and 925 ◦ C, and hornblende-out
lies between 925 and 1000 ◦ C.
5.3. Interpretation of melt compositions from
amphibolite melting experiments
The main reaction for the partial melting of amphibolite under fluid-absent conditions, at pressures where
garnet is stable is similar to:
Hbl + Pl = Grt + Cpx ± Opx + melt.
The compositions of the glasses (quenched melts)
(Table 7) are compared to the Barberton TTG rocks
in Fig. 16. They are metaluminous or marginally
peraluminous sodic granites, in many respects quite
similar in composition to some of the more felsic TTG
rocks. However, their compositions do not fall in the
trondhjemite field (Fig. 16a), where the bulk of the
Barberton TTGs plot.
73
Amp
925
1
47.44
1.07
9.59
0.21
2.00
11.55
0.10
13.57
12.39
1.50
0.58
12
5.604
0.103
3.549
0.015
0.159
2.578
0.152
1.674
1.211
0.037
0.009
16.00
39
875
6
38.41
0.94
20.72
0.10
1.85
20.96
1.23
7.59
8.00
0.14
0.05
12
5.687
0.125
3.586
0.019
0.206
2.421
0.081
1.935
1.283
0.042
0.004
15.39
44
925
5
38.67
1.13
20.69
0.16
1.86
19.69
0.65
8.83
8.15
0.15
0.02
12
5.934
0.159
3.738
0.020
0.140
2.222
0.078
2.183
1.300
0.055
0.004
15.80
50
1000
5
39.71
1.42
21.22
0.17
1.23
17.81
0.61
9.80
8.11
0.19
0.02
6
1.839
0.008
0.137
0.005
–
0.295
0.007
0.696
0.657
0.065
0.005
3.71
70
875
1
53.22
0.33
3.38
0.17
–
10.19
0.24
13.51
17.75
0.97
0.11
6
1.856
0.012
0.214
0.006
0.006
0.280
0.004
0.704
0.699
0.075
0.002
3.85
72
925
4
52.04
0.43
5.09
0.19
0.22
9.40
0.13
13.25
18.30
1.08
0.05
6
1.823
0.015
0.219
0.004
0.016
0.293
0.005
0.717
0.698
0.061
0.002
3.85
71
1000
3
51.04
0.57
5.24
0.16
0.59
9.81
0.17
13.47
18.21
0.89
0.04
6
1.996
0.005
0.258
0.008
–
0.566
0.012
0.897
0.090
0.054
0.026
3.91
61
875
2
54.75
0.20
6.00
0.26
–
18.52
0.38
16.48
2.30
0.76
0.55
6
1.883
0.006
0.157
0.003
0.009
0.574
0.010
1.196
0.050
0.023
0.001
3.91
68
925
3
52.38
0.24
3.71
0.11
0.32
19.08
0.33
22.31
1.31
0.32
0.02
6
1.859
0.012
0.131
0.004
0.015
0.554
0.007
1.212
0.050
0.014
0.001
3.85
69
1000
5
52.58
0.46
3.14
0.14
0.54
18.74
0.24
22.99
1.33
0.19
0.03
Opx
Mineral
875
6
47.25
0.88
11.78
0.17
5.61
7.86
0.26
12.86
10.92
1.79
0.61
23
6.628
0.113
1.578
0.023
0.211
1.350
0.012
2.825
1.854
0.405
0.104
15.10
68
Cpx
T (◦ C)
n
SiO2
TiO2
Al2 O3
Cr2 O3
Fe2 O3
FeO
MnO
MgO
CaO
Na2 O
K2 O
23
6.522
0.092
1.907
0.018
0.574
0.915
0.031
2.646
1.617
0.478
0.108
14.91
74
Grt
xO2−
Si
Ti
Al
Cr
Fe3+
Fe2+
Mn
Mg
Ca
Na
K
cations
Mg#
higher and the Mg# lower than in most of the Barberton TTGs (Appendix A in supplementary data). The
difference in Mg# probably reflects the relatively low
melting proportion, even at 1000 ◦ C, and the stability of
Mg-rich restitic minerals in the solid residues. Given the
observed behaviour of K2 O in the melting experiments, it
is difficult to interpret this difference as related purely to
melting conditions, although K2 O/Na2 O ratios of experimental melts, from a variety of source compositions,
do increase with increasing P (Clemens et al., in press).
Most probably, the high K2 O/Na2 O in the experimental melts reveals an important compositional difference
between the Theespruit Formation amphibolite and the
actual protolith for most of the Barberton TTGs. The
absence of Opx from the near-liquidus assemblage of the
experimental TTG contrasts with the presence of small
amounts of this phase in the partial melting experiments
on the Barberton amphibolite, a further indication that
this sort of amphibolite is probably not the dominant
source rock for the Barberton TTG magmas. Given that
this composition appears to be representative of highestgrade metamafic rocks exposed in association with the
Table 8
Average major-element compositions (normalised to 100%) of the minerals formed in fluid-absent partial melting experiments (1.6 GPa) on amphibolitised Barberton greenstone AmX12-a
(see Table 7). This suggests gradual release of K during
progressive hornblende breakdown.
In terms of residual mineralogy, the results of the
partial melting experiments agree with both our nearliquidus experiments on the TTG (THE4A), and our
interpretation of the trace-element and REE compositions of the TTG rocks. At the magma source, the TTG
melts probably coexisted with sodic hornblende, augitic
clinopyroxene and almandine-pyrope garnet, with a relatively high grossular content. Collectively, these data
suggest that the TTGs were derived from a plagioclasefree garnet amphibolite or hornblende eclogite source, at
a pressure of at least 1.5 GPa. The residual source probably contained around 20% hornblende. The experimentally observed garnet proportions (15–20%) are slightly
too low to account for the observed HREE depletion
in the TTGs, and the contribution for hornblende will be
small. This confirms that 1.5 GPa is a minimum estimate
for the pressure of melting.
The major-element compositions of the glasses match
the natural rock compositions less well. In particular,
the melt K2 O/Na2 O ratios (Table 7) are significantly
74
Constraining the pressure at which near-liquidus garnet appears in the TTGs, to >1.47 GPa, is of considerable
importance in the context of the Barberton Mountain
Land. This confirms the general high pressures of magma
derivation, implied by the REE geochemistry of the
rocks, and demonstrates that the TTG magmas must
have been produced at a depth of at least 50 km. This
implies that the 3.45 Ga Archaean crust, represented by
the southern portion of the Barberton granite-greenstone
terrane, must have been at least this thick. In a modern
tectonic scenario, this would constitute close to doubling
of the thickness of typical continental crust. This is a phenomenon that, in the modern geological record, occurs
only in tectonic settings that involve collision of terranes
containing pre-existing continental crust. In the case of
the ∼3.23 Ga TTGs, this interpretation is consistent with
the conclusions reached by several other workers who
have focused on the style and age of deformation in
the belt (de Ronde and De Wit, 1994; de Ronde and
Kamo, 2000), sedimentary environments in the upper
parts of the Barberton greenstone belt sequence (Lowe
and Byerly, 1999), and the age and P–T conditions of
the high-grade metamorphism (Dziggel et al., 2002).
Thus, these magmas appear to have been coeval with
a major terrane accretion event. This event is characterised by apparent geothermal gradients in the range
of 18 ◦ C/km (Dziggel et al., 2002; Kisters et al., 2003;
Diener et al., 2005), as indicated by mineral assemblages
in rocks exposed on the southern margin of the Belt and
xenoliths in the ∼3.45 Ga TTGs. This gradient is consistent with melting of thickened crust in the temperature
range of 850–950 ◦ C, at depths of 47–53 km.
Previous experiments have shown that TTG-like
melts can be produced from metabasaltic protoliths at
temperatures of 850–1000 ◦ C (Rapp et al., 1991; Winther
and Newton, 1991). Our partial melting experiments
confirm that the rather more K-rich Barberton greenstones would produce sodic granite partial melts, but not
trondhjemites—the dominant rock type among the Barberton TTG suite. Thus, it seems unlikely that the Barberton greenstones were the actual sources of the majority
of the TTG magmas. The results imply that, at 3.45
and 3.23 Ga, less potassic (and probably older) mafic
rocks must have underlain the Barberton terrane, and
formed the source from which the TTG magmas were
extracted. The pressure of formation of the Barberton
TTG magmas (at least 1.47 GPa) suggests that geothermal gradients during TTG formation were of the same
order at those during the metamorphism of the greenstones (around 18 ◦ C/km; Dziggel et al., 2002; Kisters
6. Discussion
TTGs, this suggests the possibility that the TTG magmas were derived from a more sodic (lower K2 O/Na2 O)
source that is not exposed among the high-grade metamorphic rocks of the Barberton greenstone belt.
Fig. 16. (a–c) Diagrams showing the compositional fields occupied
by the Barberton TTG rocks (light grey shading) and the compositions
of the three glasses (quenched melts) produced in the partial melting
experiments on greenstone amphibolite AmX12-a. White: 875 ◦ C, mid
grey: 925 ◦ C and black: 1000 ◦ C.
75
Anhaeusser, C.R., Robb, L.J., 1983. Chemical analyses of granitoid
rocks from the Barberton Mountain Land. Geol. Soc. S. Afr. Spec.
Pub. 9, 189–223.
Armstrong, R.A., Compston, W., deWit, M.J., Williams, I.S., 1990.
The stratigraphy of the Barberton greenstone belt revisited: a single
zircon ion microprobe study. Earth Planet. Sci. Lett. 101, 90–106.
Arndt, N., et al., 1998. Were komatiites wet? Geology 26, 739–742.
Arth, J.G., Barker, F., Peterman, Z.E., Friedman, I., 1978. Geochemistry of the gabbro-diorite-tonalite-trondhjemite suite of southwest
Finland and its implications for the origin of tonalitic and trondhjemitic magmas. J. Petrol. 19, 289–316.
Atherton, M.P., Petford, N., 1993. Generation of sodium-rich magmas
from newly underplated basaltic crust. Nature 362, 144–146.
Barker, F., 1979. Trondhjemite: definition, environment and hypotheses of origin. In: Barker, F. (Ed.), Trondhjemites, Dacites and
Related Rocks. Elsevier, Amsterdam, pp. 1–12.
Bea, F., 1996. Controls on the trace element composition of crustal
melts. Transactions of the R. Soc. Edinburgh Earth Sciences 87,
33–41.
References
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.precamres.
2006.08.001.
Appendix A. Supplementary data
This paper has benefited considerably from very
detailed reviews by H. Martin and an anonymous
reviewer.
Acknowledgement
their characteristic HREE depletion, were generated at
depths of at least 50 km. This inference is supported
by previous experimental work on partial melting relations of tholeiitic amphibolites, and our own experiments on a Barberton greenstone amphibolite. Restites
are likely to have been eclogitic, with mineral assemblages of Cpx + Opx + Grt ± Hbl ± Qtz + Rt + Ap. This
study has important geodynamic implications. Firstly,
the Archaean crust must have been thickened to at least
50 km in order to generate TTG magmas. This would
explain why 3.23 Ga TTG magmatism is contemporaneous with episodes of terrane collision in the tectonic
history of the Barberton region, and suggests a similar
setting for the older TTGs, where the tectonic setting
is uncertain. Secondly, the generation of TTG magmas
must have produced a plagioclase-free eclogitic restite,
in keeping with the REE geochemistry of the Barberton
TTGs. Thirdly, the parent rocks for the bulk of the Barberton TTGs were not the Barberton greenstones but an
older? mafic terrane that was structurally overlain by the
Barberton.
et al., 2003). Thus, the findings of this study have a bearing on one of the central questions in Archaean geology,
i.e. whether the mechanisms and rates of Archaean tectonics were fundamentally different to those that shaped
younger orogenic belts (e.g. Hamilton, 1998; De Wit,
1998; Collins and Van Kranendonk, 1999).
The possibility that radiogenic heat production may
have been significantly higher in the Archaean than
at present (e.g. O’Nions et al., 1978; Davies, 1993)
is central to this debate. Higher heat production may
have produced circumstances detrimental to the development of lateral tectonics (Hargraves, 1986; Davies,
1992; Vlaar et al., 1994), favouring heat dissipation
through mantle plumes (e.g. Campbell and Griffiths,
1992; Hill et al., 1992) rather than through linear magmatic spreading centres associated with lateral plate
motions. Arndt et al. (1998) reviewed the data from various sources and argued that Mg-rich komatiite lavas in
Archaean sequences (including Barberton) are dry melts
of depleted mantle. This suggests very high magma temperatures, and an Archaean mantle much hotter than at
present (e.g. Herzberg, 1995). However, the oldest TTG
rocks documented in this study, which are characterised
by HREE-depleted signatures, are the gneisses of the
Stolzburg pluton, components of which have been dated
at 3460 ± 5 Ma. Thus, melting in crust approximately
50 km thick occurred within 10–30 Myr of the time of
komatiite volcanism in the lower Onverwacht. This close
temporal association of high-pressure crustal melting
and mantle melting supports the suggestion that all these
magmas were produced in cratonic extensional basins,
in arc-continent or intracontinental rifts (Kisters et al.,
2003; Heubeck and Lowe, 1994; Lowe, 1994; Eriksson
et al., 1994), associated with subduction zones. If this
is correct, the komatiites may not represent abnormally
high-T ultramafic magmas. Instead, liquidus depression
by H2 O-rich fluid may have allowed partial melting of
the relatively shallow mantle at only moderate temperatures (e.g. Parman et al., 1997; Grove and Parman,
2004). Thus, at this time, extensional environments seem
to have been associated with crustal domains that had
low apparent geothermal gradients in which hydrated
crustal rocks underwent high-pressure metamorphism
and partial melting. Such a setting is compatible with
the occurrence of lateral tectonic processes at this early
stage of Earth’s crustal evolution.
7. Summary
Near-liquidus experiments on a Theespruit trondhjemite show that garnet is stable at P ≥ 1.47 GPa.
This suggests that the Barberton TTG magmas, with
76
archean to modern comparisons. J. Geophys. Res. Solid Earth 95,
21503–21521.
R.A., 2002. Metamorphism of the granite-greenstone terrane
south of the Barberton greenstone belt, South Africa: and insight
into the tectono-thermal evolution of the ‘lower’ portions of the
Eriksson, K.A., Krapez, B., Fralick, P.W., 1994. Sedimentology of
Archean greenstone belts—signatures of tectonic evolution. Earth
Sci. Rev. 37, 1–88.
Foley, S., Tiepolo, M., Vannucci, R., 2002. Growth of early continental
crust controlled by melting of amphibolite in subduction zones.
Nature 417, 837–840.
Gao, S., Luo, T.-C., Zhang, B.-R., Zhang, H.-F., Han, Y.-W., Zhao,
Z.-D., Hu, Y.-K., 1998. Chemical composition of the continental
crust as revealed by studies in East China. Geochim. Cosmochim.
Acta 62, 1959–11975.
Glazner, A.F., Bartley, J.M., Coleman, D.S., Gray, W., Taylor, R.Z.,
2004. Are plutons assembled over millions of years by amalgamation from small magma chambers? GSA Today 14, 4–11.
Glikson, A.Y., 1979. Early Precambrian tonalite-trondhjemite sialic
nuclei. Earth Sci. Rev. 15, 1–73.
Goldstein, S.L., O’Nions, R.K., Hamilton, P.J., 1984. A Sm-Nd isotopic study of atmospheric dusts and particulates from major river
systems. Earth Planet. Sci. Lett. 70, 221–236.
Graphchikov, A.A., Konilov, A.N., Clemens, J.D., 1999. Biotite dehydration, partial melting, and fluid composition: experiments in the
system KAlO2 –FeO–MgO–SiO2 –H2 O–CO2 . Am. Mineral. 84,
15–26.
Grove, T.L., Parman, S.W., 2004. Thermal evolution of the Earth as
recorded by komatiites. Earth Planet. Sci. Lett. 219, 173–187.
Hamilton, W.B., 1998. Archaean magmatism and plate tectonics were
Hamilton, P.J., Evensen, N.M., O’Nions, R.K., 1979. Sm–Nd dating of Onverwacht Group volcanics, southern Africa. Nature 279,
298–300.
Hargraves, R.B., 1986. Faster spreading or greater ridge length in the
Archaean. Geology 14, 750–752.
Haskin, L.A., Wildeman, T.R., Haskin, M.A., 1968. An accurate procedure for the determination of the rare earths by neutron activation.
J. Radioanal. Chem. 1, 337–348.
Herzberg, C., 1995. Generation of plume magmas through time: an
experimental perspective. Chem. Geol. 126, 1–16.
Heubeck, C., Lowe, D.R., 1994. Late syndepositional deformation
and detachment tectonics in the Barberton Greenstone-Belt, SouthAfrica. Tectonics 13, 1514–1536.
Hill, R.I., Campbell, G.F., Davies, G.F., Griffiths, R.W., 1992. Mantle
plumes and continental tectonics. Science 256, 186–193.
Jahn, B.M., Glikson, A.Y., Peucat, J.J., Hickman, A.H., 1981. REE
geochemistry and isotopic data of Archaean silicic volcanics and
granitoids from the Pilbara Block, Western Australia: implications
for the early crustal evolution. Geochim. Cosmochim. Acta 45,
1633–1652.
Jahn, B.M., Vidal, P., Kröner, A., 1984. Multichronometric ages and
origin of Archaean tonalitic gneisses in Finnish Lapland: a case for
long crustal residence time. Contrib. Mineral. Petrol. 86, 398–408.
Johnston, A.D., Wyllie, P.J., 1988. Constraints on the origin of Archean
trondhjemites based on phase relationships of Nûk gneiss with H2 O
at 15 kbar. Contrib. Mineral. Petrol. 100, 35–46.
Kamo, S.L., Davis, D.W., 1994. Reassessment of Archaean crustal
development in the Barberton Mountain Land, South Africa, based
on U-Pb dating. Tectonics 13, 167–192.
Campbell, I.H., Griffiths, R.W., 1992. The changing nature of mantle
hotspots through time: implications for the chemical evolution of
the mantle. J. Geol. 100, 497–523.
Chappell, B.W., White, A.J.R., 1974. Two contrasting granite types.
Pac. Geol. 8, 173–174.
Clemens, J.D., 1984. Water contents of intermediate to silicic magmas.
Lithos 17, 273–287.
Clemens, J.D., 2005. Melting of the continental crust: fluid regimes,
melting reactions and source-rock fertility. In: Brown, M., Rushmer, T. (Eds.), Evolution and Differentiation of the Continental
Crust. Cambridge University Press, pp. 297–331.
Clemens, J.D., Helps, P.A., Stevens, G. Chemical variations in granitic
magmas: source-inherited or products of magmatic processes?
Proc. R. Soc. Edinburgh, Earth Sci., in press.
Clemens, J.D., Vielzeuf, D., 1987. Constraints on melting and
magma production in the crust. Earth Planet. Sci. Lett. 86, 287–
306.
Clemens, J.D., Watkins, J.M., 2001. The fluid regime of hightemperature metamorphism during granitoid magma genesis. Contrib. Mineral. Petrol. 140, 600–606.
Collins, W.J., Van Kranendonk, M.J., 1999. Model for the development
of kyanite during partial convective overturn of Archaean granitegreenstone terranes: the Pilbara Craton, Australia. J. Metamorphic
Geol. 17, 145–156.
Condie, K.C., 2005. TTGs and adakites: are they both slab melts?
Lithos 80, 33–44.
Connolly, J.A.D., 2005. Computation of phase equilibria by linear programming: a tool for geodynamic modeling and its application
to subduction zone decarbonation. Earth Planet. Sci. Lett. 236,
524–541.
Davies, G.F., 1992. On the emergence of plate tectonics. Geology 20,
963–966.
Davies, G.F., 1993. Conjectures on the thermal and tectonic evolution
of the earth. Lithos 30, 281–289.
de Ronde, C.E.J., De Wit, M.J., 1994. The tectonothermal evolution
of the Barberton greenstone belt, South Africa: 490 million years
of crustal evolution. Tectonics 13, 983–1005.
de Ronde, C.E.J., Kamo, S.L., 2000. An Archaean arc-arc collisional event: a short-lived, ca 3 Myr episode, Weltevreden area,
Barberton greenstone belt, South Africa. J. Afr. Earth Sci. 30,
219–248.
De Wit, M.J., 1998. On Archaean granites, greenstones, cratons and
tectonics: does the evidence demand a verdict? Precambrian Res.
91, 181–226.
De Wit, M.J., Armstrong, R.A., Hart, R.J., Wilson, A.H., 1987. Felsic igneous rocks within the 3.3–3.5 Ga Barberton greenstone belt:
high-level equivalents of the surrounding tonalite trondhjemite terrain, emplaced during thrusting. Tectonics 6, 529–549.
Diener, J.F.A., Stevens, G., Kisters, A.F.M., Poujol, M., 2005. Geotectonic evolution of the Tjakastad Schist belt, Barberton greenstone
belt, South Africa: A record of mid-Archaean metamorphism and
terrain exhumation. Precambrian Res. 143, 87–112.
Diener J.F.A., 2004. The tectono-metamorphic evolution of the
Theespruit Formation in the Tjakastad schist belt and surrounding areas of the Barberton greenstone belt, South Africa. M.Sc.
Thesis (unpublished), Stellenbosch University, 208 pp.
Droop, G.T.R., 1987. A general equation for estimating Fe-3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineral. Mag. 51,
431–435.
Drummond, M.S., Defant, M.J., 1990. A model for trondhjemitetonalite-dacite genesis and crustal growth via slab melting:
77
Prouteau, G., Maury, R.C., Scaillet, B., Pichavant, M., 1999. Fluidpresent melting of ocean crust in subduction zones. Geology 27,
1111–1114.
Prouteau, G., Scaillet, B., Pichavant, M., Maury, R., 2001. Evidence
for mantle metasomatism by hydrous silicic melts derived from
subducted oceanic crust. Nature 410, 197–200.
Rapp, R.P., Shimizu, N., Norman, M.D., Applegate, G.S., 1999.
Reaction between slab-derived melts and peridotite in the mantle wedge: experimental constraints at 3.8 GPa. Chem. Geol. 160,
335–356.
Rapp, R.P., Shimizu, N., Norman, M.D., 2003. Growth of early
continental crust by partial melting of eclogite. Nature 425,
605–608.
Rapp, R.P., Watson, E.B., 1995. Dehydration melting of metabasalt
at 8–32 kbar: implications for continental growth and crust-mantle
recycling. J. Petrol. 36, 891–931.
Rapp, R.P., Watson, E.B., Miller, C.F., 1991. Partial melting of amphibolite/eclogite and the origin of Archean trondhjemites and tholeiites. Precambrian Res. 51, 1–25.
Robb, L.J., 1983. The nature, origin and significance of Archaean
migmatites in the Barberton Mountain Land: a new approach in
the assessment of early crustal evolution. Geol. Soc. S. Afr. Spec.
Pub. 9, 81–101.
Robb, L.J., Anhaeusser, C.R., 1983. Chemical and petrogenetic characteristics of Archaean tonalite-trondhjemite gneiss plutons in
the Barberton Mountain Land. Geol. Soc. S. Afr. Spec. Pub. 9,
103–117.
Rollinson, H.R., 1993. Using Geochemical Data: Evaluation, Presentation, Interpretation. Longman Scientific & Technical, London,
pp. 352.
Scaillet, B., Holtz, F., Pichavant, M., 1998. Phase equilibrium
constraints on the viscosity of silicic magmas—1. Volcanicplutonic association. J. Geophys. Res. Solid Earth 103, 27257–
27266.
Sen, C., Dunn, T., 1994. Dehydration melting of a basaltic composition amphibolite at 1.5 and 2.0 GPa: implications for the origin of
adakites. Contrib. Mineral. Petrol. 117, 394–409.
Skjerlie, K.P., Patiño Douce, A.E., 2002. The fluid-absent partial
melting of a zoisite-bearing quartz eclogite from 1.0 to 3.2 GPa;
implications for melting in thickened continental crust and for
subduction-zone processes. J. Petrol. 43, 291–314.
Smithies, R.H., 2000. The Archaean tonalite-trondhjemitegranodiorite (TTG) series is not an analogue of Cenozoic adakite.
Earth Planet. Sci. Lett. 182, 115–125.
Springer, W., Seck, H.A., 1997. Partial fusion of basic granulites at 5
to 15 kbar: implications for the origin of TTG magmas. Contrib.
Mineral. Petrol. 127, 30–45.
Stevens, G., Clemens, J.D., 1993. Fluid-absent melting and the roles
of fluids in the lithosphere: a slanted summary? Chem. Geol. 108,
1–17.
Vielzeuf, D., Clemens, J.D., 1992. Fluid-absent melting of phlogopite + quartz: experiments and models. Am. Mineral. 77,
1206–1222.
Viljoen, M.J., Viljoen, R.P., 1969. An introduction to the Geology of
the Barberton granite-greenstone terrain. Geol. Soc. South Africa
Spec. Pub. 2, 9–28.
Vlaar, N.J., van Keken, P.E., van den Berg, A.P., 1994. Cooling of the
Earth in the Archaean: consequences of pressure release melting
in a hotter mantle. Earth Planet. Sci. Lett. 121, 1–18.
Wells, P.R.A., 1981. Accretion of continental crust: thermal and
geochemical consequences. Phil. Trans. R. Soc. Lond. A301,
347–357.
Kisters, A.F.M., Stevens, G., Dziggel, A., Armstrong, R.A., 2003.
Extensional detachment faulting and core-complex formation in
the southern Barberton granite-greenstone terrain, South Africa:
evidence for a 3.2 Ga orogenic collapse. Precambrian Res. 127,
355–378.
Kretz, R., 1983. Symbols for rock-forming minerals. Am. Mineral. 68,
277–279.
Kröner, A., Byerly, G.R., Lowe, D.R., 1991. Chronology of early
Archaean granite-greenstone evolution in the Barberton Mountainland, South Africa, based on precise dating by single zircon
evaporation. Earth Planet. Sci. Lett. 103, 41–54.
Kröner, A., Hegner, E., Wendt, J.I., Byerly, G.R., 1996. The oldest
part of the Barberton granitoid-greenstone terrain, South Africa,
evidence for crust formation between 3.5 and 3.7 Ga. Precambrian
Res. 78, 105–124.
Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyre Le
Bas, M.J., Sabine, P.A., Schmid, R., Sorensen, H., Streckeisen,
A., Woolley, A.R., Zanettin, B., 1989. A Classification of Igneous
Rocks and Glossary of Terms. Blackwell, Oxford, 193 pp.
Lowe, D.R., 1994. Accretionary history of the Archean Barberton
Greenstone-Belt (3.55–3.22 Ga), Southern Africa. Geology 22,
1099–1102.
Lowe, D.R., Byerly, G.R., 1999. Stratigraphy of the west-central part
of the Barberton Greenstone Belt, South Africa. Geol. Soc. Am.
Spec. Pap. 329, 1–36.
Martin, H., 1986. Effect of steeper Archaean geothermal gradient on geochemistry of subduction-zone magmas. Geology 14,
753–756.
Martin, H., 1987. Petrogenesis of Archaean trondhjemites, tonalites
and granodiorites from Eastern Finland: major and trace element
geochemistry. J. Petrol. 28, 921–953.
Martin, H., 1994. The Archean grey gneisses and the genesis of the
continental crust. In: Condie, K.C. (Ed.), Archean Crustal Evolution. Elsevier, Amsterdam, pp. 205–259.
Martin, H., Moyen, J.F., 2002. Secular changes in tonalitetrondhjemite-granodiorite composition as markers of the progressive cooling of Earth. Geology 30, 319–322.
Martin, H., Smithies, R.H., Rapp, R., Moyen, J.F., Champion, D., 2005.
An overview of adakite, tonalite-trondhjemite-granodiorite (TTG),
and sanukitoid: relationships and some implications for crustal evolution. Lithos 79, 1–24.
McDonough, W.F., Sun, S.-S., 1995. Composition of the Earth. Chem.
Geol. 120, 223–253.
Moorbath, S., 1975. Evolution of Precambrian crust from strontium
evidence. Nature 254, 395–398.
Moyen, J.-F., Stevens, G., 2005. Experimental constraints on TTG
petrogenesis: implications for Archean geodynamics. In: Benn, K.,
Mareschal, J.-C., Condie, K.C. (Eds.), Archean Geodynamics and
Environments, Geophysical Monograph Series 164, pp. 149–175.
Nakamura, N., 1974. Determination of REE, Ba, Fe, Mg, Na, and K
in carbonaceous and ordinary chondrites. Geochim. Cosmochim.
Acta 38, 757–775.
O’Nions, R.K., Evensen, N.M., Hamilton, P.J., Carter, S.R., 1978.
Melting of the mantle past and present: isotopic trace element evidence. Phil. Trans. R. Soc. Lond. A258, 547–559.
Parman, S.W., Dann, J.C., Grove, T.L., De Wit, M.J., 1997. Emplacement conditions of komatiite magmas from the 3.49 Ga Komati
Formation, Barberton Greenstone Belt, South Africa. Earth Planet.
Sci. Lett. 150, 303–323.
Petford, N., Atherton, M., 1996. Na-rich partial melts from newly
underplated basaltic crust: the Cordillera Blanca Batholith, Peru.
J. Petrol. 37, 1491–1521.
78
Xiao, L., Clemens, J.D., 2006. Origin of potassic (C-type) adakite
magmas: experimental and field constraints. Lithos, submitted for
publication.
Xu, J.F., Shinjo, R., Defant, M.J., Wang, Q.A., Rapp, R.P., 2002. Origin of Mesozoic adakitic intrusive rocks in the Ningzhen area of
east China: partial melting of delaminated lower continental crust?
Geology 30, 1111–1114.
Winther, K.T., 1996. An experimentally based model for the origin of
tonalitic and trondhjemitic melts. Chem. Geol. 127, 43–59.
Winther, K.T., Newton, R.C., 1991. Experimental melting of hydrous
low-K tholeiite: evidence on the origin of Archaean cratons. Bull.
Geol. Soc. Denmark 39, 213–228.
Wyborn, D., Chappell, B.W., James, M., 2001. Examples of convective
fractionation in high-temperature granites from the Lachlan Fold
Belt. Aust. J. Earth Sci. 48, 531–541.
Precambrian Ophiolites and Related Rocks
Edited by Martin J. van Kranendonk, R. Hugh Smithies and Vickie C. Bennett
Developments in Precambrian Geology, Vol. 15 (K.C. Condie, Series Editor)
DOI: 10.1016/S0166-2635(07)15056-8
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1
2
39
41
Chapter 5.6
3
40
42
1
12
41
43
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
11
10
9
8
7
6
5
4
13
42
* Present address: Council for Geoscience, Limpopo Unit, P.O. Box 620, Polokwane 0700, South Africa
1 In this paper, we use “Barberton Granitoid-Greenstone Terrain” (BGGT) as an encompassing term to refer to
the whole area of Archean outcrops (plutons and supracrustals), as opposed to the “Barberton belt” stricto sensu,
that refers only to the supracrustal association.
Plutonic rocks constitute a large part of Archean terranes and occur mostly in the form
of variably deformed orthogneisses. The most common plutonic rocks are a suite of sodic
and plagioclase-rich igneous rocks made of tonalites, trondhjemites and granodiorites, collectively referred to as the “TTG” suite. A large body of geochemical and experimental
data exists for TTGs, and these studies have led to the general conclusion that TTGs
are essentially melts generated by partial melting of mafic rocks, mostly amphibolites (as
the dominant melting reaction involves hornblende breakdown) within the garnet stability
field. However, the geodynamic setting for the origin of TTGs is still debated and contrasting interpretations are proposed, the most common being melting of the down-going slab
in a ‘hot’ subduction zone setting (e.g., Arth and Hanson, 1975; Moorbath, 1975; Barker
and Arth, 1976; Barker, 1979; Condie, 1981; Jahn et al., 1981; Condie, 1986; Martin, 1986,
1994, 1999; Rapp et al., 1991; Rapp and Watson, 1995; Foley et al., 2002; Martin et al.,
2005), and melting of the lower part of a thick, mafic crust in an intra-plate settings (e.g.,
Maaløe, 1982; Kay and Kay, 1991; Collins et al., 1998; Zegers and Van Keken, 2001; Van
Kranendonk et al., 2004; Bédard, 2006).
In many cases, TTGs are the oldest component of Archean cratons. They generally appear as polyphase deformed gneissic complexes, commonly referred to as “grey gneisses”,
which display variable degrees of migmatization. In such units, high finite strains and
the tectonic transposition of different TTG phases obscuring original igneous contacts,
renders the recognition of original protoliths difficult and detailed geochemical studies
on individual magmatic intrusions are not possible. However, in the Barberton granitoidgreenstone terrain (BGGT)1 , many of the TTGs are characterized by weak fabrics and low
5.6-1. INTRODUCTION
Department of Geology, Geography and Environmental Science, University of
Stellenbosch, Private bag X 01, Matieland 7602, South Africa
JEAN-FRANÇOIS MOYEN, GARY STEVENS,
ALEXANDER F.M. KISTERS AND RICHARD W. BELCHER*
TTG PLUTONS OF THE BARBERTON
GRANITOID-GREENSTONE TERRAIN, SOUTH AFRICA
14
43
42
43
2
Chapter 5.6: TTG Plutons of the Barberton Granitoid-Greenstone Terrain, South Africa
strain intensities, therefore allowing detailed study of their intrusive relationships, original
compositions and comprehensive petrogenesis.
TTGs from the BGGT range in age from ca. 3.55 to 3.21 Ga and the relationship between the greenstone belt and the surrounding TTG “plutons” is complex. The apparent
domal pattern of TTG gneisses in tectonic contact with the overlying supracrustal greenstone belt is actually an oversimplification. In fact, each of the “plutons” has its own,
distinct emplacement and deformational history (summarized in Table 5.6-1), with some of
the “plutons” corresponding to relatively simple magmatic intrusive bodies, whereas others
are composite units with complex and protracted emplacement and structural histories and
are not really “plutons” in the classical sense. Likewise, the TTGs also have distinct petrological and geochemical natures, and while they all broadly belong to the “TTG” group,
are actually petrologically and geochemically complex. Such a diversity points to different
petrogenetic histories related to different geodynamic settings.
The TTGs of the BGGT can be divided in to at least two “sub-series”: (i) a “low-Sr”,
commonly tonalitic sub-series; and (ii) a “high-Sr”, commonly trondhjemitic sub-series.
In most Archean cratons, tonalites and trondhjemites are typically associated in highly
strained grey gneiss complexes, which are tectonically interleaved on a mm- to dm-scale, to
such a degree that it gives the impression that both lithologies reflect only minor differences
in terms of petrogenetic processes. In contrast, in the BGGT tonalites and trondhjemites
occur as distinct intrusive bodies with well-defined margins and intrusive contact relationships. This allows their petrogenetic evolution to be studied independently from one
another.
In this paper, we demonstrate that the tonalitic and trondhjemitic bodies reflect two fundamentally different magma types, with different origins and evolutions. We propose that
the two distinct TTG “sub-series” of the BGGT could reflect the results of two geodynamic
environments important in the formation of Archean TTG’s, namely formation at the base
of a thickened crust, and derivation from a subducting slab.
5.6-2. GEOLOGICAL SETTING
The BGGT formed between ca. 3.51 and 3.11 Ga.2 Although supracrustal rocks (lavas
and sediments) from the belt itself yield a relatively continuous spread of ages from 3559±
27 Ma (Byerly et al., 1996; Poujol et al., 2003) to 3164 ± 12 Ma (Armstrong et al., 1990;
Poujol et al., 2003), the BGGT predominantly assembled during three or four discrete
tectono-magmatic events (Poujol et al., 2003) at 3.55–3.49, 3.49–3.42, 3.255–3.225 and
3.105–3.07 Ga (see also Lowe and Byerly, this volume).
The first two events (3.49–3.55 and 3.42–3.49 Ga) are well represented in the Ancient Gneiss Complex to the east (Kröner, this volume). However, in the BGGT proper,
>3.42 Ga rocks are restricted to the high-grade “Stolzburg domain” (Fig. 5.6-1) (Kisters et
2 Ages indicated in millions of years (Ma) correspond to actual, measured ages with reference and error, while
dates given in billions of years (Ga) refer to generalized time intervals.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
17
16
15
14
15
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
4
Characteristics and emplacement features
31
Age (Ma)
32
Surface (km2 )
33
∼150
34
Pluton
Tonalitic orthogneisses, forming a mappable,
relatively homogeneous unit in the Ngwane
gneisses
35
Tsawela gneisses (in the ACG)
3455 ± 3 (York et al., 1989) to
3436 ± 6 (Kröner et al., 1993)
∼320
Composite pluton, dominated by coarse
grained, leucocratic trondhjemite Syn- to post
tectonically emplaced as a laccolith into the
greenstone belt
Polyphased gneiss domain, emplaced during a
ca. 50 Ma period, made of a variety of
mutually intrusive, diversely deformed phases
Dark, coarse-grained, amphibole bearing
tonalite. Probably emplaced as a
sub-concordant laccolith into the greenstone
belt
780
Nelshoogte pluton
∼160
3) 3.23–3.21 Ga generation
Kaap Valley pluton
Badplaas gneisses
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
aid: 15056 pii: S0166-2635(07)15056-8 docsubty: REV
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
c Interlayered with the Ngwane gneisses. Exact extension poorly known.
Kröner (this volume).
b Gneissic unit, intruded by younger plutons. Probably not only made of orthogneisses, contains some metasedimentary components (Hunter et al., 1978). See
a With inherited zircons dated at 3702 ± 1 Ma (Kröner et al., 1996).
ACG = Ancient Gneiss Complex. See Kröner et al. (this volume). Surfaces are derived using GIS from the map of Anhaeusser (1981).
3290–3240 Ma (Kisters et al., 2006);
Poujol (pers. comm.)
3229 ± 5 (Tegtmeyer and Kröner,
1987); 3227 ± (Kamo and Davis,
1994); 3223 ± 4 and 3226 ± 5 (Layer
et al., 1992); 3226 ± 14 (Armstrong et
al., 1990)
3236 ± 1 (de Ronde and Kamo, 2000);
3212 ± 2 (York et al., 1989)
42
30
29
28
Tonalitic to trondhjemitic orthogneisses,
interlayered with metasediments
34
∼2500
3683 ± 10 (Kröner et al., 1996) to
3213 ± 10 (Kröner et al., 1993)b
Tonalitic to trondhjemitic orthogneisses,
interlayered with metasediments
43
Characteristics and emplacement features
18
Age (Ma)
19
Surface (km2 )
3553 ± 4 to 3490 ± 4 (Kröner et al.,
1996)
20
Pluton
15
3540 ± 3 (Kröner et al., 1996)a
Table 5.6-1. Main field characteristic and ages of Barberton TTG plutons
1) ca. 3.5 Ga generation
Steynsdorp pluton
<1
Foliated plutons, transposed contact with
greenstones. Two gneissic units (tonalite and
granodiorite)
Fine grained gradioriorite intrusive into
Steynsdorp
∼80
3445 ± 3 (Kröner et al., 1991);
3431 ± 11 (Dziggel et al., 2002);
3460 ± 5 (Kamo and Davis, 1994)
Leucocratic trondhjemite, medium to coarse
grained. Intrusive (with intrusive breccias and
dyke swarm) into the greenstone belt, contact
deformed during the 3.2 Ga events
The same
Vlaakplats granodiorite
(intrusive in the Steynsdorp
pluton)
Elements of the Ngwane
gneisses (in the ACG of
Swaziland)
42
2) ca. 3.45 Ga generation
Stolzburg pluton
Theespruit pluton
3443 ± 4 (Kamo and Davis, 1994);
3440 ± 5 (Kröner et al., 1991);
3437 ± 6 (Armstrong et al., 1990)
No published age – probably similar to
Theespruit and Stolzburg
3448 ± 4 (Kamo and Davis, 1994)
3683 ± 10 (Kröner et al., 1996) to
3213 ± 10 (Kröner et al., 1993)
The same
3.6
∼2500
>90
Small southern plutons
(Theeboom, Eerstehoek, . . .)
Doornhoek
Elements of the Ngwane
gneisses (in the ACG of
Swaziland)
5.6-2. Geological Setting
43
42
41
40
39
38
37
36
35
33
32
31
27
26
25
24
23
22
21
20
19
18
17
16
14
13
12
11
10
9
8
7
6
5
4
3
2
1
3
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
9
11
10
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
18
17
36
38
40
6
37
39
41
5
38
40
42
39
41
43
1
40
42
23
43
42
41
40
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
5.6-2.1. >3.42 Ga Accretion of the BGGT
39
3 “Terrane” (or “block”) is used in this paper to describe a “fault-bounded geological entity with distinct
tectonostratigraphic, structural, geochronological and/or metamorphic characteristics from its neighbors (in the
sense of Coney et al., 1980)” (Van Kranendonk et al., 1993), as opposed to “terrain”, which simply refers to a
geographical region or area with no particular tectonic or genetic meaning.
The >3.5 Ga event is represented by the mafic and felsic volcanics of the Theespruit Formation (Lowe and Byerly, 1999, this volume, and references therein), which are coeval
with the emplacement of the ca. 3.55–3.50 Ga Steynsdorp pluton (Kröner et al., 1996).
Little information is available regarding the geological context of their formation.
The 3.42–3.49 Ga event corresponds to the formation of the Komati, Hooggenoeg and
Kromberg Formations of the Onverwacht Group (Lowe, 1999b; Lowe and Byerly, 1999,
this volume, and references therein), which are mostly located in the lower-grade (upper
plate of Kisters et al., 2003) portions of the Songimvelo and Steynsdorp terranes. These
three formations are dominantly mafic to ultramafic lavas, with subordinate felsic volcanic
rocks and cherts. At the contact between the Hooggenoeg and Kromberg Formations, the
ca. 3.44–3.45 Ga “H6” unit (Kröner and Todt, 1988; Armstrong et al., 1990; Kröner et al.,
1991a; Byerly et al., 1996) is nearly synchronous with the intrusion of TTG plutons in the
Stolzburg domain (Theespruit, Stolzburg, and the minor plutons to the South defined by
Anhaeusser et al., 1981). The H6 unit is a thin (few tens of meters) unit of dacitic lava
flows and shallow intrusives (geochemically regarded as the extrusive equivalents of the
22
21
20
19
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
41
43
Fig. 5.6-1. (On previous page.) Geological map of the southwestern part of the Barberton Greenstone
Belt and surrounding TTG plutons (BGGT). Left: map modified after Anhaeusser et al. (1981). See
text and Table 5.6-1 for comments and references. Top right: location map. Bottom right: Structural
sketch indicating the position of the main terranes and structures. While the “Songimvelo block” of
Lowe (1994) includes part of the Barberton Greenstone Belt, and the adjacent ca. 3.45 Ga plutons in
the south, the latter are separated from the former by the Komatii fault, leading to the identification
of a distinct “Stolzburg terrane” (Kisters et al., 2003; Kisters et al., 2004; Diener et al., 2005; Diener
et al., 2006; Moyen et al., 2006) corresponding to the amphibolite-facies portion of the Songimvelo
terrane. The main structure is the Inyoni–Inyoka fault system, separating the western (Kaap Valley
block) from the eastern domain (Steynsdorp and Songimvelo blocks, including Stolzburg terrane).
Note that the “Onverwacht Group” on both sides of the Inyoka fault actually corresponds to rocks
with different stratigraphy and of contrasting ages: 3.3–3.25 Ga to the west, and 3.55–3.3 Ga in
the east. Furthermore, the details of the stratigraphic sequences on both sides cannot be correlated,
suggesting that the two parts of the belt evolved independently, prior to the accretion along the Inyoka
fault (Viljoen and Viljoen, 1969a; Anhaeusser et al., 1981,1983; de Wit et al., 1992; de Ronde and
de Wit, 1994; Lowe, 1994; Lowe and Byerly, 1999; Lowe et al., 1999; de Ronde and Kamo, 2000).
42
al., 2003; Moyen et al., 2006; Stevens and Moyen, this volume), which corresponds to the
high-grade, “lower” portions of both the “Steynsdorp and Songimvelo terranes3 ” (Lowe,
1994, 1999; Lowe and Byerly, 1999).
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
7
TTG plutons; de Wit et al., 1987) and clastic sediments and conglomerates. This suggests
that some topography existed at that stage. The first, well constrained deformation event
affecting the belt (D1 ) (Lowe et al., 1999) also occurred at about the same time and is
interpreted to represent the development of an active margin (oceanic arc) (Lowe, 1999b;
de Ronde and Kamo, 2000; Lowe and Byerly, this volume, and references therein) at ca.
3.45 Ga.
Following the D1 event, the Mendon Formation was deposited in the Stolzburg domain
(Songimvelo and Steynsdorp blocks) in the east (Lowe, 1999b), and the Weltvreden Formation in the western terranes, from ca. 3.42 to 3.25 Ga. Based on studies of the volcanic
and sedimentary units, a period of quiescence (rift/intracontinental setting) is suggested
(Lowe, 1999).
5.6-2.2. Main Orogenic Stage at 3.25—3.21 Ga
The main, “collision” stage (D2–5 ), occurred between 3.25 and 3.21 Ga. Evidence for an
accretionary orogen is presented elsewhere (Lowe and Byerly, this volume; Stevens and
Moyen, this volume), and is thus only briefly summarized here. D2 corresponds to the
amalgamation of the various sub-terranes that make up the belt, with the major suture
zone corresponding to the Inyoni–Inyoka fault system (Fig. 5.6-1). Despite the apparently
continuous stratigraphy across the fault, the sequences on both sides cannot be correlated
(Lowe, 1994, 1999; Lowe et al., 1999; Stevens and Moyen, this volume). The D2 event
is shortly followed by deposition (syn D3 ) and deformation (D4 and D5 ) of the <3.22 Ga
(Tegtmeyer and Kröner, 1987) Moodies Group conglomerates and sandstones.
The most likely sequence of events for this stage is:
– from ca. 3.25 to 3.23 Ga, syn-tectonic (D2a ) deposition of the felsic volcanics and clastic
sediments of the Fig Tree Group, probably resulting in the development of a volcanic
arc in what is now the terrane west of the Inyoni–Inyoka fault system (Lowe, 1999b; de
Ronde and Kamo, 2000; Kisters et al., 2006). The Badplaas gneisses were also emplaced
into the western terrane during this period.
– accretion of the two terranes along the Inyoni–Inyoka fault system at ca. 3.23 Ga (D2b ).
This was accompanied by high-pressure, low- to medium-temperature metamorphism
of the eastern, Stolzburg domain (Dziggel et al., 2002; Diener et al., 2005; Moyen et
al., 2006), especially along the fault system, interpreted as a suture zone (Stevens and
Moyen, this volume).
– collision was immediately followed at ca. 3.22–3.21 Ga by extensional collapse of the
orogenic pile (Kisters et al., 2003), leading to nearly isothermal exhumation of the highpressure rocks of the Stolzburg domain along detachment faults (Diener et al., 2005;
Moyen et al., 2006) and the emplacement of TTG plutons (Nelshoogte and Kaap Valley
plutons). The extension collapse roughly corresponds to the D3 event of Lowe (1999b)
and was synchronous with deposition of (at least part of) the Moodies Group in small,
discontinuous, fault-bounded basins (Heubeck and Lowe, 1994a, 1994b). This was immediately followed by diapiric exhumation of the lower crust, and steepening of the
fabrics.
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
8
– Late, ongoing deformation (D4 –D5 ) resulted in strike-slip faulting and folding of the
whole sequence (including the Moodies Group). Some late to post-tectonic plutons (e.g.,
3215 ± 2 Ma Dalmein pluton; Kamo and Davis, 1994), crosscut all ca. 3.23–3.21 Ga
structures.
5.6-2.3. Later Events at ca. 3.1 Ga
At ca. 3.1 Ga, a final orogenic event (not named in Lowe’s (1999) terminology) resulted
in intraplate compression (Belcher and Kisters, 2006a, 2006b) and widespread melting at
different crustal levels (Belcher et al., submitted). This led to the emplacement of voluminous, sheeted potassic batholiths and the development of a network of syn-magmatic
shear zones that affected the older “basement” (Westraat et al., 2004). The volumetrically
dominant intrusions in the BGGT (Fig. 5.6-1) were emplaced at this time (Maphalala and
Kröner, 1993; Kamo and Davis, 1994) and are represented by the Piggs’ Peak batholith
(east of the BGGT and in Swaziland), Nelspruit batholith (in the north), and the Mpuluzi/Lochiel and Heerenveen batholiths (in the south). Collectively, these rocks are mostly
leucogranites, granites and granodiorites, associated with minor monzonites and syenites,
and commonly referred to as the “GMS” (granites/granodiorites, monzonites and syenites/syenogranites) suite (Yearron, 2003). Although the GMS suite formed, at least in part,
from partial melting of rocks compositionally similar to the 3.5–3.2 Ga rocks of the BGGT
(Belcher et al., submitted), the TTG “basement” observed in outcrop across the terrain was
unaffected by this melting event.
5.6-3. TTG PLUTONS OF THE BGGT
5.6-3.1. Geology and Field Relationships of TTG Plutons
TTGs of the BGGT belong to three main generations, corresponding to the three geological
events outlined above (Table 5.6-1).
– The ca. 3.55–3.50 Ga TTGs, represented by the Steynsdorp pluton (Robb and Anhaeusser, 1983; Kröner et al., 1996), contain a pervasive solid-state gneissosity and
occurs mostly as banded gneisses. The protolith of these gneisses is tonalitic (Kisters
and Anhaeusser, 1995b; Kröner et al., 1996), although a granodioritic component, possibly related to the remelting of older tonalites or trondhjemites (see below), is also
recorded. The Steynsdorp pluton outcrops in a domal antiform (Kisters and Anhaeusser,
1995b), and the contact with the enveloping supracrustals of the Theespruit Formation
is tectonic.
– The ca. 3.45 Ga (syn-D1 ) TTGs are represented by a number of intrusive bodies in the
Stolzburg terrane, located to the south of the main part of the Barberton Greenstone Belt
(Viljoen and Viljoen, 1969d; Anhaeusser and Robb, 1980; Robb and Anhaeusser, 1983;
Kisters et al., 2003; Moyen et al., 2006). The two most prominent and better defined
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-3. TTG Plutons of the BGGT
9
intrusions are the Stolzburg and Theespruit plutons. Together with the smaller Doornhoek pluton, these plutons intruded the supracrustal rocks of the belt. Further south,
several smaller plutons or domains are recognized and form a complex pattern of TTG
gneisses and greenstone remnants, partially transposed and dismembered by ca. 3.1 Ga
shear zones. These are the Theeboom, Eerstehoek, Honingklip, Weergevonden “cells”
and “plutons” of Anhaeusser et al. (1981), see also Robb and Anhaeusser (1983). To
the west, the Stolzburg terrane is bounded by the Inyoni shear zone, which is the southern extent of the Inyoni–Inyoka fault system. Rocks predominantly from the Stolzburg
pluton are foliated and transposed in this shear zone, in a ∼500 m wide area. To the
north, it is truncated by the extensional detachment corresponding to the Komati Fault
(Kisters et al., 2003). Within the terrane, the plutons preserve clear intrusive relationships with the surrounding greenstones (Fig. 5.6-2(a)) (Kisters and Anhaeusser, 1995a;
Kisters et al., 2003), although the terrane as a whole (granitoids and country rocks) were
deformed during D3 exhumation (Kisters et al., 2003; Diener et al., 2005, 2006; Stevens
and Moyen, this volume). The nature of the preserved contacts, which clearly cut across
amphibolite-facies foliations (Fig. 5.6-2(a)), the presence of a network of surrounding
dykes, and the existence of simultaneous, cogenetic extrusive rocks all suggest that the
Stolzburg pluton (and the other plutons of the terrane/domain) intruded under brittle conditions, at relatively shallow crustal levels (Kisters and Anhaeusser, 1995a). All the ca.
3.45 Ga plutons are composed predominantly of medium- and/or coarse-grained leucotrondhjemites (Robb and Anhaeusser, 1983; Kisters and Anhaeusser, 1995a; Yearron,
2003). Minor dioritic dykes are also observed (Yearron, 2003), especially in the margins
of the plutons, and in the complex inter-pluton areas.
– The 3.29–3.21 Ga group (syn-D2 and D3 ) of plutons is more composite, and occurs
along the northern and southwestern margins of the Barberton Greenstone Belt (Viljoen
and Viljoen, 1969d; Anhaeusser and Robb, 1980; Robb and Anhaeusser, 1983).
• In the south, the 3.29–3.24 Ga (Kisters et al., 2006, Poujol, pers. comm.) Badplaas gneisses (and probably the apparently similar Rooihoogte gneisses, west of the
3.1 Ga Heerenveen batholith) are composed of two main suites: an older, coarsegrained leuco-trondhjemitic component that underwent solid-state deformation; and a
younger, multiphase intrusive component, made up a variety of typically finer-grained
trondhjemites. In proximity to the Inyoni shear zone, most of these intrusions are syntectonic. the Batavia pluton, of coarse-grained, leucocratic, porphyritic trondhjemite,
is syntectonic in the central part of the Inyoni Shear Zone (Anhaeusser et al., 1981).
Further away from the shear zone, the trondhjemites form either irregularly shaped,
discontinuous, stockwork-like breccias or small (100 m – 5 km) plugs and intrusions.
The long-lived emplacement of the Badplaas pluton, and its composite nature, makes
it unique in the BGGT.
• Further north, the composite 3.23–3.21 Ga Nelshoogte pluton (Anhaeusser et al.,
1981, 1983; Robb and Anhaeusser, 1983; Belcher et al., 2005) is dominated by coarsegrained leuco-trondhjemites that are intruded by amphibole-tonalites, particularly
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
42
10
41
43
43
42
41
40
42
Fig. 5.6-2. Field appearance of the various type of TTG rocks around Barberton Greenstone Belt.
(a) Lit-par-lit and cross-cutting intrusive relations between the 3.45 Ga Stolzburg pluton and amphibolites of the Theespruit Formation. (b) Brecciation of Onverwacht Group amphibolites by the
3.23–3.21 Ga Nelshoogte pluton.
43
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
37
39
12
34
38
40
11
35
39
41
36
40
42
39
40
41
43
42
41
40
39
38
37
36
37
41
43
Fig. 5.6-2. (Continued.) (e) Trondhjemitic orthogneisses in the 3.23–3.21 Ga Nelsghoogte pluton.
(f) Hornblende-bearing tonalites of the 3.23–3.22 Kaap Valley pluton. Microgranular mafic enclaves
(Didier and Barbarin, 1991), as seen in this photo, are common. Coin for scale in photos (c)–(f).
42
38
43
Fig. 5.6-2. (Continued.) (c) Banded tonalitic gneisses of the 3.55–3.50 Ga Steynsdorp pluton.
(d) Leucocratic coarse-grained trondhjemites from the 3.45 Ga Stolzburg pluton. The Stolzburg pluton shows a pronounced vertical rodding not seen in this image, which is taken on a plane orthogonal
to the stretching lineation. Coin for scale in photos (c)–(f).
42
along the northern and northeastern margin of the pluton. The pluton was intruded
during regional folding, probably as a laccolith, and lit-par-lit intrusive relationships,
as well as smaller-scale brecciation with the surrounding greenstone wallrocks, are
preserved (Belcher et al., 2005) (Fig. 5.6-2(b)). This is again suggestive of relatively
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
13
shallow level of emplacement. The domal map pattern reflects late stage folding and
steepening of the syn-emplacement, initially flat fabrics.
• The large, 3.23–3.22 Ga Kaap Valley pluton along the northern margin of the Barberton Greenstone Belt is, for the most part, made up of coarse-grained, biotiteamphibole tonalite (Robb et al., 1986), with minor occurrences of amphibole-tonalite
(biotite free).
5.6-3.2. The Pristine Character of Barberton TTGs
A rather unique feature of Barberton TTGs is that they represent a group of well-defined,
distinct intrusions. Apart from the Badplaas gneisses, they do not constitute a heterogeneous complex of orthogneisses (grey gneisses) like many other TTG complexes that often
are polyphase, high strain, transposed and often migmatitic, or even poly-migmatitic, orthogneisses. Although the TTGs around Barberton are all technically gneisses, in the sense
that they underwent solid-state deformation after their emplacement (most likely related
to the 3.2 Ga D2 –D3 event: Kisters and Anhaeusser, 1995a; Kisters et al., 2003; Belcher
et al., 2005), they still commonly contain original magmatic and emplacement features
and textures. In the 3.45 Ga plutons, for instance, deformation occurred 150–200 My after their emplacement and is marked by strong, subvertical rodding (D3 ), corresponding
to pure coaxial stretching. However, strain intensities are low enough to allow magmaticlooking textures to be preserved, at least in planes perpendicular to the lineation. Likewise, emplacement-related features and intrusive contacts are also occasionally preserved
(Kisters and Anhaeusser, 1995a) and deformation did not result in transposition and development of a gneissic fabric, but rather limited textural overprinting along the margin of
the plutons and the immediately surrounding wallrocks. Consequently, unlike many grey
gneisses terrains in the world, their composition has not been altered by partial melting.
The Barberton TTGs thus preserve their true magmatic compositions and present a very
good example to study the origin and evolution of TTG magmas (s.s.), as opposed to the
geochemistry of grey gneisses complexes (even though these are dominated by TTGs).
Some banded grey gneisses are known in the BGGT. However, they represent part of
well-constrained high strain zones, corresponding to 3.23–3.21 Ga (e.g., the Inyoni Shear
Zone: Kisters et al., 2004) or 3.1 Ga (e.g., the Weltverdiend Shear Zone: Westraat et al.,
2004) tectonic events. Within these zones, the complex orthogneisses are very similar to
any other grey gneiss complex in the world, being characterized by transposed, high strain
fabrics, amphibolite enclaves, etc. However, the field relationships of the components are
obvious, and they clearly formed by deformation of the plutonic rocks and supracrustal
remnants.
A second, equally important point is that Barberton TTGs are relatively high-level, intrusive bodies. Although no quantitative data is available, the emplacement mode of all
plutons (except, perhaps, part of the Badplaas gneisses) are suggestive of emplacement
in the middle or upper crust, under brittle conditions (Kisters and Anhaeusser, 1995a,
1995b; Kisters et al., 2003, 2004; Belcher et al., 2005). Barberton TTG plutons are not
migmatitic domes, with liquids and solids still intermingled, nor are they lower-crustal
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
14
diatexitic bubbles rising diapiricaly. Rather, they are “clean” (purely or mostly magmatic
liquids), high-level plutons emplaced sometimes as syn-tectonic magmas and sometimes
deformed during subsequent events. As the TTG melts were probably generated at depths
greater than 10–12 kbar (see below), this implies that they were emplaced far (at least 15–
20 km above) from their source. The present outcrop level is entirely disconnected from
the melting domain.
5.6-3.3. Petrology and Mineralogy
Although few Archean geologists would refer to them in this way, TTGs are I-type granites
(Chappell and White, 1974) and belong to a calc-alkaline series (Le Bas et al., 1986; Le
Maître, 2002). However, they do show significant differences with typical, modern calcalkaline lavas or arc-related I-type granitoids.
Two main rock-types are represented in the TTG rocks of the BGGT (Fig. 5.6-2(c–e)):
5.6-3.3.1. Leucocratic biotite trondhjemite
Several types of trondhjemites are observed in Barberton TTG plutons (Yearron, 2003).
They range from fine- to coarse-grained rocks, with occasional porphyritic varieties; all
have similar mineralogy, dominated by plagioclase (oligoclase to andesine; 55–65%),
quartz (15–20%), biotite (5–15%) and microcline (∼10%). Accessory minerals are apatite, allanite and (magmatic) epidote, with secondary chlorite, sericite and saussurite. It is
worth noting that the name “trondhjemite” is synonymous with “leuco-tonalite” and should
be used only for rocks with less than 10% mafic minerals, less than 10% alkali feldspar,
and more than 20% quartz (Le Maître, 2002). Obviously, some samples of this rock type do
not strictly fit the definition, and are “tonalites”, “granodiorites”, or even “(leuco-) quartz
monzonites”; however, the name “trondhjemites” fits most of the samples and is retained
for convenience.
5.6-3.3.2. Hornblende tonalite
Hornblende tonalites are found in the Kaap Valley pluton and the northern margin of the
Nelshoogte pluton (Robb et al., 1986; Yearron, 2003; Belcher et al., 2005). Smaller, pluglike and isolated tonalitic intrusions also occur in the southern TTG-gneiss terrain around
the Schapenburg schist belt (Anhaeusser et al., 1983; Stevens et al., 2002) and along the
western margin of the large Mpuluzi batholith (Westraat et al., 2004). They are dominated
by plagioclase (oligoclase to andesine; ∼60%), interstitial quartz (10–20%), and subhedral
hornblende (∼15%) with minor biotite and microcline and accessory allanite and ilmenite.
Secondary chlorite and epidote develop at the expense of hornblende. In places, more mafic
dioritic enclaves are common, displaying the same mineral assemblage as the hornblende
tonalites, but in different proportions. With less than 20% quartz, some of the “tonalites”
are technically leuco-quartz-diorites (in IUGS terms) (Le Maître, 2002).
In contrast to the trondhjemites, tonalites are absent from the ca. 3.45 Ga group. They
are found in parts of the Steynsdorp pluton, and represent the latest (syn- to post-tectonic)
stages of the ca. 3.29–3.21 Ga group.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
5.6-4. Geochemistry
15
5.6-3.3.3. Minor components
– Mafic dykes are observed as a minor component of many of the plutons, most commonly
in the ca. 3.2 Ga plutons. Some dioritic dykes also occur in the ca. 3.45 Ga TTGs,
especially along the margins of the individual plutons. The diorites have a mineralogy
similar to the “wall rock” trondhjemite or tonalites (Yearron, 2003), but with different
mineral proportions (60% plagioclase, 15% quartz, 10% each biotite and amphibole,
some microcline – Yearron (2003)). Gabbroic dykes are also reported, but not described
(Yearron, 2003) in the Nelshoogte and Kaap Valley plutons.
– Felsic dykes range from leucocratic versions of the TTGs, to plagiogranites, to porphyries and aplites, or pegmatites. All point to some degree of in-situ differentiation,
probably fluid assisted, or they are related to the later, ca. 3.1 Ga, event. Collectively
however, their volume is too small to represent more than local processes.
Clearly, in the typical “grey gneiss” terrains of most Archean provinces, these diverse
components would be interleaved and transposed with the dominant trondhjemites or
tonalites, resulting in some difficulty in explaining the scatter of compositions of these
gneissic units. This is not the case in the relatively low strain BGGT.
1
2
3
4
5
6
7
8
9
10
11
12
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
7
Theespruit
8
6
4
3
2
1
28
27
22
16
15
17
14
16
13
12
11
10
11
10
5
5
73.15
0.24
15.46
1.36
0.02
0.46
2.37
5.32
1.55
0.07
16
Stolzburg
ca. 3.45 Ga
9
3.55–3.50 Ga
70.34
0.25
16.04
2.01
0.03
1.10
2.85
5.74
1.56
0.08
1.05
0.29
0.15
0.45
THE6B
Theespruit
Trondhjemite
High-Sr
Yearron 2003
70.99
0.26
15.25
1.73
0.05
0.73
1.61
7.37
1.95
0.06
0.98
0.27
0.18
0.50
THE4A
Theespruit
Trondhjemite
High-Sr
Yearron 2003
73.41
0.22
14.12
1.62
0.04
0.46
1.43
6.45
2.20
0.07
0.89
0.26
0.11
0.22
STZ11
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
75.11
0.14
14.04
1.26
0.02
0.18
1.13
3.56
4.54
0.04
0.91
0.34
0.10
0.22
STZ10
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
65.67
0.46
17.70
3.75
0.06
1.82
3.89
4.43
1.99
0.24
1.09
1.28
0.08
0.32
STY4B
Steynsdorp
High-K
(Low-Sr)
Yearron 2003
1.07
0.45
0.22
0.88
STY1
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
Steynsdorp
STY2A
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
Major elements (wt%)
SiO2
70.52
0.37
TiO2
Al2 O3
15.93
FeOt
2.75
MnO
0.04
MgO
1.25
CaO
2.95
Na2 O
4.60
1.40
K2 O
P2 O5
0.19
LOI
Selected ratios
A/CNK
1.10
0.30
K2 O/Na2 O
0.19
CaO/AI2 O3
CaO/Na2 O
0.64
Source
Sample
Pluton
Type
Table 5.6-2. Representative analysis of Barberton TTGs, for the different plutons studied
41
13
14
15
16
17
18
19
20
21
22
23
24
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
25
26
27
28
29
30
31
32
33
34
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
26
25
24
23
21
20
19
18
15
14
13
12
9
8
7
6
4
3
2
1
35
36
37
38
39
40
41
42
43
42
5.6-3.4. Summary
The TTGs of the BGGT are spatially and temporally distinct from one another. Geographically, the ca. 3.45 Ga TTGs are in the east, and younger 3.2 Ga old rocks are in the west,
separated from one another by the Inyoni shear zone, which is interpreted to represent a suture zone during ca. 3.25–3.21 Ga orogenesis. This temporal and spatial distinction is also
recorded in the compositions of these rocks. The 3.45 Ga generation is only trondhjemitic,
whereas the 3.29–3.21 Ga plutons are both trondhjemitic and tonalitic; in this latter group,
the tonalites always represent the youngest phases, either as the slightly younger Kaap Valley pluton, or as late intrusive phases in composite plutons. The switch from trondhjemitic
(3.29–3.22 Ga) to tonalitic (3.22–3.21 Ga) compositions at ca. 3.22 Ga appears to coincide
with a change in geological regime from collision tectonics to orogenic collapse.
5.6-4. GEOCHEMISTRY
Numerous analyses of Barberton TTGs have been published (Anhaeusser and Robb,
1980, 1983; Anhaeusser et al., 1981; Robb et al., 1986; Kleinhanns et al., 2003; Yearron,
2003). Unfortunately, many are either relatively old and were not obtained with modern
mass spectrometry techniques, or samples were crushed with carbide tungsten mills, such
that the existing database, while extensive, is not particularly consistent and lacks reliable
determination for some important elements (Ni, Cr, Ta, Pb). The following discussion is
based on 314 analyses from published (see references above) and unpublished data (Table 5.6-2). Unpublished analyses were obtained in 2004–2005 at Stellenbosch University
and University of Capetown. Major elements and some traces were analyzed by XRF,
whereas most traces including REE have been analysed in-situ by LA-ICP-MS on glass
beads, following the procedure described in Belcher et al. (submitted).
43
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
43
42
41
40
39
38
37
36
33
32
31
30
27
26
25
24
21
20
19
18
14
13
12
STZ11
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
15
9
8
7
29
28
29
28
23
22
23
22
17
16
17
16
11
3
2
1
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
42
41
40
39
38
37
36
33
32
31
30
29
27
26
25
24
21
20
19
17
16
15
17
14
16
13
12
23
22
23
22
9
8
7
6
4
3
2
1
18
18
13
11
12
11
10
5
7
6
5
38.4
70.8
19
26.8
24
11.7
19.8
28
THE6B
Theespruit
Trondhjemite
High-Sr
Yearron 2003
11.6
21.6
90
THE4A
Theespruit
Trondhjemite
High-Sr
Yearron 2003
Theespruit
10
117
STZ11
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
18
110
STZ10
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
13.7
25.8
30
34
34
28
28
Stolzburg
STY4B
Steynsdorp
High-K
(Low-Sr)
Yearron 2003
ca. 3.45 Ga
44
3.55–3.50 Ga
STY1
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
21.9
45.7
10.0
1.60
0.54
1.36
0.18
36
35
33
30
29
27
16.0
27.8
8.6
1.89
0.62
0.87
24.1
49.2
11.4
2.33
0.54
1.08
12.0
15.7
2.88
0.50
2.21
0.33
12.1
2.47
0.84
2.78
0.38
0.56
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
116.4
0.54
0.46
0.09
0.48
0.06
1.31
0.18
73.0
41.6
2.5
1.05
17.0
35.4
2.2
1.13
16.2
66.3
2.1
0.61
31.7
35.1
2.6
1.49
13.8
48.5
6.0
0.99
8.1
44
STY2A
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
Steynsdorp
35
LaN
YbN
Eu/Eu*
(La/Yb)N
REE(ppm)
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Selected ratios
Sr/Y
Nb/Ta
Source
Sample
Pluton
Type
43
13
12
11
6
8
5
7
6
5
13.7
18
22.1
36.5
19
15.0
47.6
45.7
517.6
18.7
98.4
6.7
24
15.0
2.0
44.0
51.3
623.8
7.0
127.1
4.6
211.5
25
2.0
32.0
54.0
586.0
5.0
103.0
311.6
30
34.0
44.0
551.0
5.0
95.0
374.0
31
48.3
73.0
40.5
6.9
97.2
111.7
166.2
5.5
128.4
7.3
480.0
THE6B
Theespruit
Trondhjemite
High-Sr
Yearron 2003
4
THE4A
Theespruit
Trondhjemite
High-Sr
Yearron 2003
Theespruit
10
STZ10
Stolzburg
Trondhjemite
High-Sr
Yearron 2003
35
34
35
34
Stolzburg
STY4B
Steynsdorp
High-K
(Low-Sr)
Yearron 2003
ca. 3.45 Ga
STY1
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
3.55–3.50 Ga
STY2A
Steynsdorp
Tonallite
Low-Sr
Yearron 2003
100.5
586.6
13.3
205.6
11.1
575.4
3.3
316.5
6.9
Steynsdorp
Sample
Pluton
Type
Source
Trace elements (ppm)
Sc
V
38.1
Cr
32.0
Ni
20.4
Cu
20.8
Zn
65.0
Ga
Ge
As
Rb
66.4
Sr
495.5
Y
11.3
Zr
183.8
Nb
7.2
Sb
Ba
247.2
Hf
Ta
Pb
Th
U
5.6-4. Geochemistry
43
42
41
40
39
38
37
36
33
32
27
26
21
20
15
14
10
9
4
3
2
1
17
43
42
41
40
39
38
37
32
31
26
25
21
20
15
14
9
8
4
3
2
1
43
42
41
40
39
38
37
36
21
20
19
Nelshoogle
11
10
9
8
NLG5
Nelshoogle
Trondhjemite
Low-Sr
Yearron
2003
7
6
5
4
SKV20
Kaap
Tonallite
Low-Sr
Anhaeusser
and Robb
1983
35
34
33
32
31
30
29
28
27
26
25
24
23
18
17
16
15
14
13
12
2
1
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
42
41
40
39
38
37
36
35
34
33
32
30
29
28
27
26
25
24
23
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
17-99/128
Kaap
Tonallite
Low-Sr
This work
5
4
3
2
1
20
74.6
56.3
14.9
54.7
17.3
SKV20
Kaap
Tonallite
Low-Sr
Anhaeusser
and Robb
1983
6
2.0
20.0
Nelshoogle
13.7
41.4
NLG5
Nelsnoogle
Trondhjemite
Low-Sr
Yearron
2003
27.8
50.2
NLG21A
Nelsnoogie
Trondhjemite
Low-Sr
Yearron
2003
BTV16A
Badplaas
Trondhjemite
Low-Sr
Yearron
2003
23.2
83.0
NLG15
Nelsnoogie
Tonallite
Low-Sr
Yearron
2003
BDP8C
Badplaas
Trondhjemite
High-Sr
Yearron
2003
14.4
61.6
NLG14C
Nelsnoogle
Tonallite
Low-Sr
Yearron
2003
33.1
50.0
272.0
6.0
71.0
9
8
6
5
4
48
570
222.0
40.0
547.8
37.7
555.2
6.5
96.4
3.5
39.1
572.0
9.0
103.0
3.8
269.1
191
39.4
823.1
7.2
177.6
1.7
237.0
2.6
0.2
4.7
1.7
0.5
253.1
64
15.7
53.5
460.9
14.1
80.7
3.5
45
266.9
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
85
57.4
312.3
15.2
154.6
5.3
114
132.3
33
113.9
1.9
21
143.4
31
4.7
63.6
253.6
2.1
42.5
1.0
0.9
297.2
1.3
0.1
25.8
2.5
1.3
4-7-05B
Badplaas
High-K
(low-Sr)
Yearron
2003
9.6
5.6
10.1
35.6
22.8
0.2
1.6
34.7
576.4
1.8
78.9
2.0
0.9
216.9
1.8
0.2
12.1
1.6
0.4
124
9.2
22.9
25.1
24.6
313
9.1
22
3.23–3.21 Ga
Badplass
Cr
31.4
Ni
7.9
Cu
15.3
Zn
63.7
Ga
29.8
Ge
1.7
As
11.1
Rb
45.0
Sr
799.0
Y
3.9
Zr
141.7
Nb
2.6
Sb
1.4
Ba
180.8
Hf
3.1
Ta
0.2
Pb
13.2
Th
2.3
U
0.6
Selected ratios
Sr/Y
205
Nb/Ta
12.5
Sample 4-7-05A
4-7-13B
Pluton Badplaas Badplaas
Type
MeltMeltdepleted
depleted
This work This work
Source
43
8.1
75.7
0.79
0.23
0.28
0.71
64.51
0.51
16.11
3.95
0.06
2.71
4.56
6.45
1.47
0.19
17-99/128
Kaap
Tonallite
Low-Sr
This work
3
60.51
0.57
16.31
4.10
0.06
3.27
4.32
5.31
1.20
0.19
3.47
NLG21A
Nelshoogie
Trondhjemite
Low-Sr
Yearron
2003
73.91
0.16
14.34
1.18
0.05
0.51
1.91
6.27
1.61
0.06
0.91
0.23
0.26
0.81
NLG15
Nelshoogie
Tonallite
Low-Sr
Yearron
2003
0.92
0.26
0.13
0.30
NLG14C
Nelshoogle
Tonallite
Low-Sr
Yearron
2003
4-7-05B
Badplaas
High-K
(low-Sr)
Yearron
2003
1.03
0.25
0.19
0.57
BTV16A
Badplaas
Trondhjemite
Low-Sr
Yearron
2003
4-7-13B
Badplaas
Meltdepleted
This work
0.87
0.24
0.30
0.93
15.0
BDP8C
Badplaas
Trondhjemite
High-Sr
Yearron
2003
22
Badplass
4-7-05A
Badplaas
Meltdepleted
This work
0.82
0.33
0.32
0.97
39.1
70.11
0.26
16.06
2.72
0.05
1.05
3.01
5.28
1.34
0.11
1.05
0.33
0.22
0.78
88.6
63.13
0.62
17.33
4.04
0.04
2.63
5.18
5.56
1.31
0.17
75.47
0.12
14.47
0.39
0.05
0.34
0.64
3.36
3.76
0.08
1.66
1.06
0.21
0.20
0.65
86.9
63.37
0.46
15.15
5.00
0.10
4.48
4.78
4.91
1.61
0.14
73.47
0.19
15.19
1.05
0.05
0.45
1.39
4.57
1.80
0.10
1.61
1.34
1.12
0.04
0.19
58.7
68.67
0.46
16.39
3.40
0.06
1.25
3.55
4.54
1.52
0.14
1.27
0.39
0.09
0.30
15.1
71.21
0.19
16.56
1.83
0.03
0.62
3.32
5.09
1.09
0.08
2.4
1.8
10.5
3.23–3.21 Ga
Sample
Pluton
Type
Source
Major elements (wt.%)
SiO2
70.67
0.28
TiO2
Al2 O3
16.53
FeOt
2.02
MnO
0.06
MgO
1.04
CaO
3.59
3.71
Na2 O
K2 O
0.99
0.13
P2 O5
LOI
1.68
Selected ratios
A/CNK
1.21
0.27
K2 O/Na2 O
CaO/AI2 O3
0.22
0.97
CaO/Na2 O
Trace elements (ppm)
Sc
3.3
V
14.5
5.6-4. Geochemistry
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
19
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
7
3
2
1
43
42
41
40
39
38
37
36
21
20
Nelshoogle
22
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
35
34
33
32
31
30
29
28
27
26
25
24
23
15.0
4.01
1.06
2.61
7.4
1.48
0.51
1.48
0.24
14.5
33.9
12.6
9.8
18.5
8.0
3.34
0.50
1.02
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
44.9
3.6
1.08
12.3
14.8
31.7
3.9
15.1
2.99
0.96
2.48
0.34
1.76
0.34
0.89
0.13
0.80
0.11
17-99/128
Kaap
Tonallite
Low-Sr
This work
1
13.0
2.49
0.65
2.24
0.36
0.60
2
8.9
1.50
0.48
1.15
0.07
43.8
29.6
2.7
0.83
10.8
3
Badplass
3.23–3.21 Ga
Sample
Pluton
Type
0.33
0.05
Source
1.14
19.3
1.5
1.06
12.9
4-7-05A
4-7-13B
4-7-05B BDP8C
BTV16A
NLG14C
NLG15
NLG21A
NLG5
SKV20
Badplaas Badplaas Badplaas Badplaas
Badplaas
Nelsnoogle Nelsnoogie Nelsnoogie
Nelsnoogle
Kaap
MeltMeltHigh-K Trondhjemite Trondhjemite Tonallite
Tonallite
Trondhjemite Trondhjemite Tonallite
depleted
depleted
(low-Sr) High-Sr
Low-Sr
Low-Sr
Low-Sr
Low-Sr
Low-Sr
Low-Sr
This work This work Yearron Yearron
Yearron
Yearron
Yearron
Yearron
Yearron
Anhaeusser
2003
2003
2003
2003
2003
2003
2003
and Robb
1983
1.03
0.11
40.4
5.2
1.01
7.8
6.4
13.3
0.15
0.03
41.6
4.7
0.85
8.7
13.3
25.7
11.1
18.3
1.8
6.6
1.74
0.43
1.12
0.14
0.47
0.11
0.19
0.04
1.03
0.04
46.0
0.7
1.12
67.5
13.7
24.6
0.30
0.04
33.5
4.7
0.96
7.2
15.2
23.4
REE (ppm)
La
15.2
Ce
27.2
Pr
2.8
Nd
9.0
Sm
1.79
Eu
0.78
Gd
1.04
Tb
0.19
Dy
0.84
Ho
0.18
Er
0.60
Tm
Yb
0.64
Lu
0.15
37.6
1.4
1.84
27.3
12.4
20.0
2.0
7.1
1.16
0.50
0.61
0.11
0.41
0.07
0.16
45.9
2.9
1.75
15.8
LaN
YbN
Eu/Eu*
(La/Yb)N
“Type” refers to the discussion paragraphs; “high-Ca mafic” and “high-K” refer to the “non-TTG” facies, whereas “tonalite” and “trondhjemite” correspond to
the two main types of TTGs. Major elements are in wt%, traces in ppm. L.O.I.: loss on ignition. A/CNK: molecular ratio Al/(Ca+Na+K). LaN , YbN : normalized
REE values (Nakamura, 1974). Eu/Eu* = EuN /(0.5 × (SmN + GdN )) is a measure of the “depth” of the Eu anomaly (a negative Eu anomaly corresponds to
Eu/Eu* < 1).
22
13
12
11
15
14
13
12
11
1
14
16
2
15
17
1
16
18
3
17
19
2
18
20
4
19
21
3
20
22
5
21
23
4
22
24
6
23
25
5
24
26
7
25
27
6
26
28
8
27
29
7
28
30
8
29
31
10
30
32
9
31
33
9
32
34
10
33
35
5.6-4. Geochemistry
43
42
41
40
39
38
37
36
34
Fig. 5.6-3. Major elements features of Barberton TTGs: (a) Total alkali vs silica (TAS) diagram
(Cox et al., 1979); (b) FMA (Fe-Mg-alkali) diagram (Irvine and Baragar, 1971); (c) Silica-potassium
diagram (Peccerillo and Taylor, 1976), showing the potassic rocks, probably formed by remelting of
earlier TTGs, with a distinctive vertical trend in this diagram. The three diagrams allow us to characterize the TTG rocks as belonging to a sub-alkaline (a), low-to-medium-K (b), and calc-alkaline
(c) series. Symbols: analyses are grouped according to their chemistry (Section ??); colours indicate whether the sample belongs to a low- or high-Sr sub-series, whereas the symbol differentiates
between “true TTGs” (tonalites or trondhjemites), high-K2 O rocks, and “melt-depleted” samples.
35
36
37
38
39
40
41
42
43
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
21
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-4. Geochemistry
Fig. 5.6-4. Normative feldspar triangle (O’Connor, 1965), for the studied units. Each unit is plotted in one individual panel to allow comparison. Same caption as Fig. 5.6-3. The fields are labeled only in the first panel: Tdj, trondhjemite; Ton, tonalite; Grd, granodiorite; QMz,
quartz-monzonite; Gr, granite.
23
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
39
41
24
40
42
5.6-4.1.1. Major elements
The two rock types identified above (tonalites and trondhjemites) display some differences in terms of major elements contents. The tonalites are silica-poorer (typically 62–
68 wt%), whereas the trondhjemites are more felsic (typically 70–75 wt%). Accordingly,
the tonalites are richer in FeO and MgO and marginally poorer in Na2 O, K2 O and CaO.
Both the tonalites and the trondhjemites belong to a sub-alkaline, calc-alkaline series
(Fig. 5.6-3(a,b)) with their volcanic equivalents being “soda-rhyolites”, dacites and mi-
43
42
41
40
39
38
37
36
35
34
33
32
31
30
41
43
Fig. 5.6-5. Harker plots for major elements for Barberton TTG. FeOt = total iron as FeO; Mg#
= molecular ratio 100 Mg/(Mg+Fe); A/CNK = molecular ratio Al/(Ca+Na+K). Symbols as in
Fig. 5.6-3.
42
5.6-4.1. Common Characteristics
43
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
26
34
36
26
27
35
37
25
28
36
38
5.6-4. Geochemistry
29
37
39
35
36
37
38
39
5.6-4.1.2. Trace elements
To some degree, all TTGs of the BGGT present comparable features. Like all TTGs, they
have low concentrations of compatible transition elements (Ni, Cr, V), relatively low HFSE
contents (Ti, Zr, Hf) and moderately high LILE and fluid-mobile elements contents (Rb,
Ba, Th). LILE/HFSE ratios are higher than in modern arc-related magmas (Pearce, 1983).
One of the most characteristic features of the TTGs from the BGGT is the high Sr contents
(typically 500–1000 ppm) and associated low Y values (average 7.8 ppm), which confers a
43
42
41
40
39
38
37
36
35
34
33
32
31
30
30
38
40
40
41
32
39
41
31
33
40
42
Fig. 5.6-6. Harker plots for selected trace elements and ratios for Barberton TTGs. Eu/Eu* =
EuN /(0.5(SmN +GdN )), normalization values after Nakamura (1974). Symbols as in Fig. 5.6-3.
34
41
43
scatter and differences between samples or sample groups, which will be used to further
subdivide the TTG rocks in several sub-series.
42
nor andesites (for the tonalites). Most of the samples reviewed belong to a medium-K
series (Fig. 5.6-3(c)), but a significant part of the 3.23–3.21 Ga group, especially in the
Badplaas unit, belongs to a low-K series. Most of BGGT rocks belong to the high-Al TTG
group of Barker and Arth (1976). In a (normative) Ab-An-Or diagram (O’Connor, 1965)
(Fig. 5.6-4), the data plots mostly in the trondhjemite field (leucocratic facies), extending
into the granite field, or in the tonalite and granodiorite fields (hornblende tonalite). All
these characteristics are typical of most TTG rocks (Martin, 1994). In Harker diagrams
(Fig. 5.6-5), most elements display a compatible behavior, with all samples plotting along
similar trends. However, Al2 O3 , K2 O and Na2 O show a different pattern, with much more
43
42
43
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
27
37
39
41
28
28
38
40
42
27
29
39
41
43
5.6-4. Geochemistry
30
40
42
34
35
36
37
38
39
40
41
32
41
43
31
33
42
high Sr/Y ratio (typically around 100). In Harker-type diagrams (Fig. 5.6-6), it is possible
to identify several groups with different trends; diagrams such as SiO2 vs. Sr/Y or La/Yb,
for instance, clearly show that the tonalites (and some of the trondhjemites) on one hand,
and most of the trondhjemites on the other hand, define contrasting sub-horizontal and
sub-vertical trends, respectively. To some degree, the same grouping can be observed with
most of the other elements, Sr being the most discriminating.
REE patterns display high LREE (LaN = 40–60) and low HREE (YbN < 5) contents,
corresponding to rather fractionated REE patterns ((La/Yb)N = 10–25) that lack Eu anomalies. This is lower than most TTGs, which have LaN values of ∼100, and (La/Yb)N of
35–40 (Martin, 1994) (Fig. 5.6-7).
43
42
43
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
34
36
38
30
35
37
39
29
36
38
5.6-4. Geochemistry
37
39
1
38
40
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
5
4
3
2
6
39
41
41
Fig. 5.6-7. REE patterns of Barberton TTGs (normalized to chondrite after Nakamura (1974). In all
diagrams, the thick grey line corresponds to the TTG average of Martin (1994). Note the opposition
between the low-Sr plutons (Kaap Valley, Nelshoogte) that mostly plot above the average for HREE,
and the high-Sr plutons (Stolzburg, Theespruit) that mostly plot below. Also note the important scatter for the composite Badplaas gneisses.
7
40
42
5.6-4.2. Distinct Geochemical Types
5.6-4.1.3. Discrimination diagrams
In discrimination diagrams (Pearce et al., 1984), TTGs always fall in the “VAG” (volcanic
arc granites) field, reflecting their low HFSE, Y and Yb contents. However, no genetic
implication should be drawn from this observation. Indeed, geotectonic diagrams like these
are based on compilations of analyses of rocks from known tectonic settings. Using these
diagrams for Archean rocks (in a genetic sense) implies that Archean magmas formed in
similar contexts and via similar processes to modern magmas. In the case of the Archean,
this carries the implicit assumptions that: (1) modern-style plate tectonics operated during
the Archean; and (2) that its modalities (thermal regimes, rock type presents, etc.) were
the same as present-day situations. Both assumptions are far from proven, and therefore
geotectonic “discrimination” diagrams should not be used in pre-Phanerozoic times, as
pointed out in the original paper by Pearce et al. (1984).
The most classical discrimination diagrams used for the interpretation of TTG petrogenesis, however, reflect their REE, Sr and Y contents (Fig. 5.6-8). In Sr/Y vs. Y and La/Yb
vs. Yb diagrams (Martin, 1986, 1987, 1994), TTGs plot along the Y-axis, distinct from
modern, calc-alkaline magmas (and most I-type granites). However, Barberton TTGs tend
to cluster in the lower-left corner of both diagrams, close to, or in the overlap area between,
the two fields.
These observations imply the existence of a phase with a high partition coefficient (Kd )
for Y and the heavy REE at some stage during TTG petrogenesis. Among the common minerals, only garnet and to a lesser degree amphibole (Rollinson, 1993; Bédard, 2006) have
adequate Kd values, implying that either (or both) coexisted as solid phases with the magma
at some stage of its evolution, and were not entrained in the plutons as observed now.
Another significant feature of Archean TTGs, in general, is their variable, but typically
low, Nb/Ta ratios (Kambers et al., 2002; Kleinhanns et al., 2003; Moyen and Stevens,
2006). However, the data presented herein, being relatively incomplete, does not document
this well.
41
43
43
42
41
40
42
In addition to the shared characteristics presented above, it is possible to identify several
sub-series, with distinct geochemical signatures; these sub-series are most clearly differentiated by their Sr contents and their K2 O/Na2 O nature, defining:
(1) a K2 O poor sub-series, either “low Sr” or “high Sr”;
(2) a K2 O rich sub-series;
(3) a high Eu/Eu* , very low K2 O, “melt-depleted” sub-series.
Fig. 5.6-8. Some trace elements characteristics of Barberton TTGs. In both diagrams, the grey field is
the TTG field and the stippled field delineates modern calc-alkaline magmas (Martin, 1994). Symbols
as in Fig. 5.6-3. The right-hand side panel has a double scale, both in ppm and in normalized values
(Nakamura, 1974).
43
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-4. Geochemistry
31
5.6-4.2.1. Low and high Sr sub-series
Regardless of their petrologic nature (tonalite or trondhjemite), the TTGs of the BGGT can
be classified in to sub-series on the basis of their position in a SiO2 -Sr diagram (Fig. 5.69(a)).
Although the absolute values of Sr abundances are comparable in the two sub-series,
the low SiO2 group defines a lower Sr trend for a given SiO2 value (Fig. 5.6-9(a)). A very
similar observation is made by Champion and Smithies (this volume) for TTGs in the
Pilbara Craton, although TTGs from the BGGT have collectively higher Sr values than
Pilbara rocks (the Barberton low-Sr group has Sr levels comparable to the Pilbara high-Sr
rocks). In the Pilbara, Champion and Smithies (this volume) observed that Al2 O3 contents
also reflect this difference, with the high-Sr sub-series also being high-Al. This is only
partially supported by our data: most of our samples (low and high-Sr together) have the
same Al contents as, and plot together with, the Pilbara high-Sr group. On the other hand,
a R1 -R2 diagram (de la Roche et al., 1980), which takes into account most major elements,
clearly differentiates between the two sub-series (Fig. 5.6-9(b)). The high-Sr sub-series is
also somewhat more sodium-rich (and with higher Na2 O/CaO) than the low-Sr series. The
low-Sr rocks also tend to have higher Y contents (or, rather, a larger range of Y values for
a given SiO2 content), giving them lower Sr/Y ratios.
The high-Sr rocks are mostly trondhjemites, with 68% < SiO2 < 75%; rare samples
have lower SiO2 contents and are tonalitic. The low-Sr group, in contrast, comprises both
tonalites (with SiO2 < 68%) and trondhjemites (68% < SiO2 < 77%). In an O’Connor
(1965) normative diagram, the high-Sr group occupies almost exclusively the trondhjemite
field, whereas the low-Sr sub-series plot in the tonalite, trondhjemite and granodiorite
fields. In the low-Sr group, the tonalites and trondhjemites are clearly differentiated, not
only by their SiO2 contents, but also by the flatter trend of the tonalites in SiO2 -Sr binary
diagrams at ca. 600 ppm Sr (Fig. 5.6-9(a)).
The subdivision of TTGs into a low- and high-Al sub-series is not new, and was proposed more than 30 years ago (Barker and Arth, 1976). All our samples, however, (like
most of the world’s TTGs, see for instance Martin (1994)) belong to what would be a highAl2 O3 series, according to these definitions; the (relatively subtle) differences we observe
reflect subdivisions of Barker’s high-Al group.
5.6-4.2.2. High K sub-series
High K2 O felsic rocks form a minor component of, for example, the Steynsdorp and Badplaas units. These rocks are not always possible to identify in the field. In some cases, like
the granodioritic phases of the Steynsdorp pluton, their K-feldspar rich nature is immediately obvious. Sometimes, however, they are macroscopically indistinguishable from the
more common trondhjemites and they can only be identified geochemically. Such “potassic” facies have relatively high K2 O/Na2 O (>0.5). They plot in the medium to high-K
fields of a SiO2 -K2 O diagram (Fig. 5.6-3(c)) at ca. 70% SiO2 , with no clear trends, and
are mostly granites (in an O’Connor diagram) (Fig. 5.6-4). In the R1 -R2 diagram (Fig. 5.69(b)), they clearly plot below (lower R2 values) the other rock types. They have low Y
(mostly <10 ppm), Yb (<1 ppm), and sometimes a slight negative Eu anomaly. Most of
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
42
32
41
43
43
42
41
40
39
38
42
Fig. 5.6-9. (a) SiO2 vs Sr and (b) R1 –R2 diagrams for Barberton TTGs. R1 and R2 are the cationic
parameters of de la Roche (1980). The two sub-series define two distinct trends: a low-Sr trend
corresponding to the lower SiO2 tonalitic facies, but including some of the higher-SiO2 rocks; and
a high-Sr trend mostly corresponding to the high SiO2 trondhjemites. The low-Sr trend also evolves
from higher R2 values. The high K2 O series has low R2 values and the “melt-depleted” samples have
high R1 . The trend of Pilbara TTGs is from Champion and Smithies (this volume).
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-4. Geochemistry
33
the “high K” rocks belong to the low-Sr series (Fig. 5.6-9(a)), with very low Sr contents
(<250 ppm). In Sr/Y vs Y or La/Yb vs Yb diagrams, they are virtually indistinguishable
from the “normal” (sodic) TTGs, although they tend to plot “below” the field of ordinary
TTGs in a Sr/Y vs Y diagram (Fig. 5.6-8), reflecting lower contents of both Sr and Y. The
high-K2 O rocks also have high LILE contents (Rb, Ba, U, of course K) and are quite similar to the “enriched TTGs” reported in the Pilbara Craton by (Champion and Smithies, this
volume).
5.6-4.2.3. “Melt-depleted” samples
Some rocks with uncommon geochemistry are found in the Badplaas gneisses that belong to a very low K2 O series (around 1% K2 O or less, Fig. 5.6-3(c)), but are not very
rich in Na2 O (∼4%). They are Al2 O3 -enriched and correspondingly have high, to very
high, A/CNK ratios (1.2–1.4), consistent with their chlorite-rich mineralogy (the chlorite
is probably a secondary mineral, but reflects an Al-rich composition whatever the primary
minerals were; garnet is occasionally observed). Thee rocks have high R1 values and plot
to the right of most other samples in a R1 -R2 diagram (Fig. 5.6-9(b)). They tend to have
high Sr, and high Sr/Y ratios (up to >400), the highest in the whole database (Fig. 5.68). Finally, they have small positive Eu anomalies (Fig. 5.6-7). Geochemically, they are
therefore the opposite of the high K2 O group.
All these features, combined with the long-lived, multiphase nature of the Badplaas
gneisses, suggest that they represent a melt-depleted facies; i.e., restitic rocks out of which
some melt has been extracted, represented by the high K2 O rocks in the Badplaas gneisses.
Whether the melt extraction reflects the 3.29–3.22 Ga evolution of the Badplaas domain,
or rather the later, ca. 3.1 Ga formation of the nearby Heerenveen batholith (Belcher et al.,
submitted), is uncertain.
5.6-4.3. Summary and Subdivision in the Different Plutons
On a geochemical basis, four groups of rocks were identified; high-K2 O rocks, low-Sr
and high-Sr “true TTGs”, and “melt-depleted” gneisses of the Badplaas unit. The “meltdepleted” rocks are not, strictly speaking, magmas (although their origin is related to
magmatic evolution). The three other types can be distinguished by devising a “Sr”
vs K2 O/Na2 O diagram. Plotting Sr in a diagram can be misleading, as this parameter is
strongly correlated to SiO2 ; Sr values alone do not allow differentiation between low and
high SiO2 series, which are distinguished by Sr contents at a given SiO2 level. To overcome this problem, we calculate a new parameter, Sr, that represents the distance of
an analysis from a reference line in a SiO2 -Sr diagram (Fig. 5.6-9(a)). Here, this line is
taken as the dividing line between low and high-Sr sub-series, allowing straight-forward
interpretations: low-Sr sub-series rocks have negative Sr, while high Sr sub-series samples have positive Sr. In our case, the reference line follows the following equation:
Srref = 4621 − 57.14 · SiO2 , and therefore Sr = Sr − Srref for each individual sample.
K2 O/Na2 O is also correlated to SiO2 , and ideally it would be possible to calculate a
“K” parameter in the same way. The benefit would, however, be minimal, since the
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
34
range of K2 O/Na2 O values between the normal TTGs and the high-K2 O group exceeds
the variations within a group.
The K2 O/Na2 O vs. Sr diagrams presented in Fig. 5.6-10, therefore, allow us to distinguish the main groups.
The ca. 3.55 Ga Steynsdorp pluton is made of two components, a low-Sr tonalitic to
trondhjemitic facies, and a high-K2 O unit. Both are now interleaved, but have been identified in the field.
The ca. 3.45 Ga group (Stolzburg and Theespruit plutons) appear as a largely homogeneous population. It is primarily composed of high-Sr trondhjemites, although some
examples of low-Sr tonalites are found in the database.
The 3.29–3.21 Ga group is more complex. The older Badplaas gneisses encompass samples belonging to both the high and low-Sr sub-series, together with high-K2 O samples and
melt-depleted rocks. The Nelshoogte and Kaap Valley plutons are both made up of low-Sr
trondhjemites and low-Sr tonalites; the trondhjemites are dominant in the Nelshoogte pluton, whereas the tonalites form most of the Kaap Valley pluton – a fact somehow obscured
in Fig. 5.6-10 by sampling bias. In other words, the 3.29–3.21 Ga group probably records
a transition from high-Sr trondhjemites, to low-Sr trondhjemites, to low-Sr tonalites.
5.6-4.4. Isotopes
Whole rock Sr-Nd isotopic data have been published on Barberton TTGs (Barton et al.,
1983; Kröner et al., 1996; Yearron, 2003; Sanchez-Garrido, 2006). Unfortunately, only
one study (Sanchez-Garrido, 2006) gives combined Sr and Nd data for the studied samples.
Collectively, 18 Nd isotopic analyses and 61 Sr data are published, but only 5 combined SrNd analyses. However, combining the (independently) published data allows us to define
the probable range of compositions (Fig. 5.6-11).
TTG plutons mostly have isotopic characteristics close to the bulk Earth, with εNd values between +4 and −3 and εSr between −7 and +5 (ISr values of 0.6995 to 0.701).
This is a commonly observed feature of Archean TTGs (e.g., Bickle et al., 1983; Martin,
1987; Peucat et al., 1996; Whitehouse et al., 1996; Bédard and Ludden, 1997; Berger and
Rollinson, 1997; Liu et al., 2002; Whalen et al., 2002; Stevenson et al., 2006; Zhai et al.,
2006; Champion and Smithies, this volume). This implies that TTGs are derived from juvenile, or newly extracted sources, either the mantle itself or more probably, basalts recently
extracted from the mantle.
In the case of the Barberton TTGs, however, there is a systematic difference between the
older (3.45 Ga) and the younger (3.29–3.21 Ga) TTGs, the former having more juvenile
characteristics (high εNd and low εSr ) than the latter.
The 3.29–3.21 Ga group was possibly derived from either pre-existing rocks of the Onverwacht Group (Hamilton et al., 1979; Kröner et al., 1996), its high-grade equivalents in
the Swaziland Ancient Gneiss Complex (Kröner et al., 1993; Kröner and Tegtmeyer, 1994;
Kröner, this volume) or even the Fig Tree Group (Toulkeridis et al., 1999; Sanchez-Garrido,
2006). Alternatively, the relatively enriched signature of the 3.23–3.21 Ga generation could
reflect a composite source, including both depleted and enriched (recycled or already emplaced crust?) components.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Fig. 5.6-10. Sr vs. K2 O/Na2 O diagram. Sr is a parameter representing the distance from the low/high Sr groups dividing line; see definition
in the text. The vertical and horizontal dashed lines correspond, respectively, to the limit between “true TTGs” and “high-K2 O”, and between
low- and high-Sr groups, effectively defining 4 sub-series (although the high-K2 O, high-Sr is virtually non-represented, such that only three
sub-series really exist).
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Fig. 5.6-11. Isotopic characteristics of TTG plutons around Barberton Belt. (a) εNd vs εSr diagram (Zindler and Hart, 1986). ε values are
calculated at the age of formation of these rocks (Table 5.6-1), this diagram is therefore not drawn for a specific time. Individual analyses
(when both Sr and Nd data are available) are plotted as individual symbol, symbols corresponding here to different plutons. One Sr-Nd analysis
showing an aberrant εSr value is not plotted. When no coupled analyses are published, the box for each pluton is bounded by the extreme range
of Sr isotopic data (in X) and the extreme range of Nd isotopic data (in Y). (b) Nd isotopic evolution diagram, using the data of (Kröner et
al., 1996; Yearron, 2003; Sanchez-Garrido, 2006) for TTG plutons (individual analyses), and (Hamilton et al., 1979; Kröner and Tegtmeyer,
1994; Kröner et al., 1996) for supracrustals (grey fields). The light grey band corresponds to Onverwacht Group mafic and ultramafic lavas, the
darker grey to Onverwacht metasediments, intermediate and felsic lavas, and to their equivalents in the Ancient Gneiss Complex in Swaziland.
Depleted mantle is linearly interpolated from εNd = 0 at 4.56 Ga to εNd = +10 now (Goldstein et al., 1984). In both case, note the difference
between the ca. 3.45 Ga plutons, with isotopic characteristics intermediate between the depleted mantle and the CHUR or the Onverwacht
mafics and ultramafics, and the isotopically more evolved ca. 3.23–3.21 Ga plutons, consistent with derivation from an enriched mantle source
or from part of the Onverwacht crust. CHUR values are 143 Nd/144 Nd = 0.512638; 147 Sm/144 Nd = 0.1967; 87 Sr/86 Sr = 0.7045; 87 Rb/86 Sr =
0.0827 (Goldstein et al., 1984).
5.6-4. Geochemistry
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
36
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-5. Petrogenesis of TTG Rocks
37
On the other hand, a depleted mantle component (or basalts derived from it) must have
played at least some role in the origin of the 3.45 Ga generation. This essentially rules out
their generation by partial melting of a pre-existing old cratonic crust.
5.6-5. PETROGENESIS OF TTG ROCKS
Different hypothesis (not always mutually exclusive, but separated here for clarity) have
been proposed to account for the origin of TTG magmas, in general. The most common
are:
(1) Partial melting of mantle, either directly to generate felsic magmas (Model 1a) (Moorbath, 1975; Stern and Hanson, 1991; Bédard, 1996), or indirectly to form basaltic
or andesitic melts that subsequently fractionate amphibole ± plagioclase ± garnet
(Model 1b: Arth et al., 1978; Barker, 1979; Feng and Kerrich, 1992; Kambers et al.,
2002; Kleinhanns et al., 2003).
(2) Partial melting of crustal plagioclase + biotite ± quartz-rich lithologies (either
metagraywackes or earlier tonalites: Arth and Hanson, 1975; Kröner et al., 1993;
Winther, 1996; Bédard, 2006).
(3) Partial melting of mafic lithologies (metabasalts, either as amphibolites or eclogites),
either in intraplate conditions in the lower part of a thick oceanic or continental crust
(Smithies and Champion, 2000; Whalen et al., 2002; Bédard, 2006; Champion and
Smithies, this volume) or in a subducting slab (Arth and Hanson, 1975; Moorbath,
1975; Barker and Arth, 1976; Barker, 1979; Condie, 1981; Jahn et al., 1981; Condie,
1986; Martin, 1986; Rapp et al., 1991; Martin, 1994; Rapp and Watson, 1995; Martin,
1999; Foley et al., 2002; Martin et al., 2005).
These three hypotheses will now be briefly discussed:
5.6-5.1. TTG as Mantle Melts?
Felsic magmas can be generated directly from the mantle (Model 1a), assuming very low
melt fractions (<5%). Calc-alkaline magmas are generated from wet mantle, typically
above active subduction zones. However, both experimental (Mysen and Boettcher, 1975a,
1975b; Green, 1976; Green and Ringwood, 1977; Wyllie, 1977) and theoretical (Jahn et al.,
1984; Pearce and Parkinson, 1993; Martin, 1994; Kelemen, 2003) approaches show that,
in this case, the melts are andesitic (and potassic), rather than tonalites and trondhjemites,
as they are formed through the breakdown of potassic hydrous phases (richterite or phlogopite, Millhohlen et al., 1974; Sudo, 1988; Tatsumi, 1989; Tatsumi and Eggins, 1995;
Schmidt and Poli, 1998), with no significant amounts of garnet in the residuum.
Alternately, andesitic or basaltic magmas generated in the mantle could fractionate
amphibole ± garnet ± plagioclase and evolve towards more felsic, HREE-depleted compositions (Model 1b). This has been shown to be possible both on experimental (Alonso-Perez
et al., 2003; Grove et al., 2003) and theoretical (Kambers et al., 2002; Kleinhanns et
al., 2003) grounds. However, such a process would require large amounts (up to 75%)
of mafic-ultramafic cumulate (Martin, 1994), which are largely missing from the BGGT.
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
38
Furthermore, as pointed by Bédard (2006), the inferred parental magma – an andesite or
andesitic basalt – is an uncommon rock type in the Archean, indeed unknown in Barberton
greenstone belt both at 3.45 or 3.23 Ga.
5.6-5.2. TTGs as Melts of Pre-Existing Felsic Lithologies?
Melting of biotite (or amphibole) – plagioclase – quartz assemblages has been experimentally demonstrated to generate broadly tonalitic to trondhjemitic magmas (Gardien et al.,
1995; Patiño-Douce and Beard, 1995; Winther, 1996; Patiño-Douce, 2005). Owing to relatively potassic sources (compared to mafic or ultra-mafic sources), this process results
in the formation of magmas with a distinct geochemical signature, characterized by subvertical trends in SiO2 -K2 O diagrams, relatively high K2 O/Na2 O values and higher LILE
concentrations. In Barberton TTGs, rocks belonging to the high-K2 O group do correspond
to this description, and can be safely attributed to the melting of comparatively enriched,
relatively potassic (and probably felsic) sources. Again, this corresponds to the interpretation proposed by Champion and Smithies (this volume) for the Pilbara LILE-enriched,
“transitional TTGs”.
The nature of the felsic source, in the regional context, is uncertain. In the Badplaas
unit, the presence of “melt-depleted” gneisses with matching geochemical characteristics
suggests that, at least for this unit, the high-K2 O rocks proceed from partial melting of already emplaced TTGs. A similar explanation is likely for the Steynsdorp pluton, where the
“potassic” unit represents a sizable volume. On the other hand, in all other studied intrusions, high-K2 O rocks are a minor, very uncommon type, precluding important remelting
of the TTGs. High-K2 O rocks could represent late melt mobilization during emplacement;
alternately, they could reflect minor source heterogeneities. Indeed, the supracrustal pile of
the BGB contains, even in the Onverwacht Group, minor sediment layers or felsic lavas
(see above), and is not a perfectly homogeneous pile of basalts. During melting, such heterogeneities would yield potassic, LILE-enriched melts in small volumes. Most of them
would be diluted and assimilated into the dominant TTG component, but it is possible that
small magma batches are somehow preserved and retain their geochemical characteristics.
At high melt fractions, melting of TTG-like sources would of course generate melts
whose composition would be very close to the source, to the point of becoming hardly
distinguishable (Bédard, 2006). Bulk recycling of a tonalitic/trondhjemitic crust would,
therefore, produce a continuum of compositions, from low melt fraction, high-K2 O liquids, to higher melt fraction, tonalitic to trondhjemitic liquids. Whereas this is more or less
observed in the Badplaas unit, such a continuum is lacking from all other plutons, suggesting that bulk recycling of older TTG gneisses typically was not an important process in
their generation.
5.6-5.3. TTGs as Melts of Mafic Lithologies?
The most common hypothesis for TTG genesis is partial melting of mafic lithologies
(metabasalts) dominated by plagioclase and amphibole. This is supported by ample experimental (reviewed in Moyen and Stevens, 2006) and geochemical (reviewed in Martin,
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-5. Petrogenesis of TTG Rocks
39
1994) evidence. The major element composition of the TTGs, in general, is explained
by fluid-absent melting of plagioclase-amphibole assemblages (Rushmer, 1991; Rapp and
Watson, 1995; Vielzeuf and Schmidt, 2001; Moyen and Stevens, 2006); i.e., melting during which water was supplied by the breakdown of hydrous phases (either amphibole,
or sometimes epidote). This is an incongruent melting reaction, in which solid products
(commonly garnet and/or orthopyroxene) are generated in addition to melt. The dominant
melting reactions will be either:
(1) Amphibole + Plagioclase = Melt + Ti-oxides + Orthopyroxene ± Clinopyroxene ±
Olivine
(Beard and Lofgren, 1991; Rapp et al., 1991; Rushmer, 1991; Patiño-Douce and Beard,
1995; Rapp and Watson, 1995; Zamora, 2000; Vielzeuf and Schmidt, 2001), at pressures
below garnet stability (i.e., P < 10–12 kbar); or
(2) Amphibole + Plagioclase = Melt + Garnet + Ti-oxides ± Clinopyroxene
at higher pressures (Rapp et al., 1991; Rapp and Watson, 1995; Zamora, 2000; Vielzeuf
and Schmidt, 2001).
While the role of plagioclase accounts for the sodic nature of the melts, the presence of
mafic peritectic phases keep them leucocratic, by locking up the Fe and Mg to very high
temperatures (>1100 ◦ C). Trace element characteristics are largely due to the presence of
garnet in the residuum (either as a preexisting phase, or as a peritectic product), implying
melting at pressures above 10–12 kbar. Therefore, there is now a large consensus on the
fact that TTGs are the products of partial melting of mafic lithologies in the garnet stability
field.
Despite this large consensus, details of the processes involved are debated. Several parameters can affect the melt composition of the melt, and a large part of the debate focuses
on “which set of parameters better matches all characteristics of TTGs”.
The geodynamic environment of melting is also debated. Indeed, metabasites can reach
melting conditions within the garnet stability field, in several conceivable scenarios:
– A commonly proposed model for the generation of Archean TTGs is that they were
generated within a subducting slab of oceanic crust. Under presumably hotter Archean
conditions, slab melting was probably favored over slab dehydration, resulting in the relatively easy and widespread generation of TTG melts, rather than dehydration of the slab
causing mantle wedge fertilization and eventually leading to the formation of andesites
(Martin, 1986, 1987, 1994). Such a process is observed in the formation of adakites,
which are in many respects modern-day analogues of Archean TTGs (Martin, 1999;
Martin et al., 2005). However, this view has been increasingly criticized in the recent
years, on several grounds:
• Modeling of the thermal structure of the slab is inconclusive. There is no definitive
proof that slab melting could be a widespread or universal phenomena in the Archean
(review in Bédard, 2006, paragraph. 2.3), but there is no definitive proof of the opposite, either. Actually, such models depend critically on too many unconstrained
parameters, such as the potential mantle temperature, mantle composition, thicknesses
of oceanic and continental lithosphere and crustal thicknesses, to be able to provide
better than semi-quantitative answers.
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
40
• It has been suggested that the volume of magmas formed by subduction-type
processes is not able to generate the large TTG batholiths observed in Archean terranes (Whalen et al., 2002; Bédard, 2006). Such calculations, however, rely on many
rather unconstrainable assumptions (thickness of the subducting slab, 3D shape and
volume of the TTG intrusions, precise timing of events, etc.). For instance, in the
BBGT, many of the younger 3.2 Ga TTG plutons have recently been suggested to
represent upfolded, possibly relatively thin laccoliths, rather than voluminous diapiric
bodies (Kisters et al., 2003; Belcher and Kisters, 2005). This makes the issue surrounding whether enough magma can be generated or not somewhat less pertinent.
• In the case of slab melting, felsic TTG liquids would form at relatively low melt fractions (Moyen and Stevens, 2006), raising issues surrounding how they are extracted
from the source, and their ascent mechanism through a hot mantle wedge. In the case
of modern adakites, high Ni, Cr, and Mg contents are ascribed to melt-mantle interactions during ascent (Kelemen, 1995; Smithies, 2000; Martin and Moyen, 2002; Martin
et al., 2005). Evidence for similar processes in Barberton TTGs is, however, cryptic
(see below).
Collectively, it seems likely that Archean slab melting could, and did, occur, but
was not the universal process as previously assumed.
– Over an active subduction zone, in underplated basalts undergoing subsequent remelting
(Gromet and Silver, 1983; Petford and Atherton, 1996). Assuming the overriding plate
was thick enough, underplating of basalts would occur at a sufficient depth to be in
garnet stability field, and subsequent remelting would indeed generate TTG magmas.
– At the base of a thick crust, either continental or oceanic, either away from any plate
boundary (e.g., oceanic plateau: Maaløe, 1982; Kay and Kay, 1991; Collins et al., 1998;
Zegers and Van Keken, 2001; Van Kranendonk et al., 2004; Bédard, 2006; Champion
and Smithies, this volume) or over tectonically thickened crust (de Wit and Hart, 1993;
Dirks and Jelsma, 1998). Many such models involve delamination of the dense lower
crust, resulting in heating of the mafic stack and pervasive melting of its base, accompanied by diapiric rise of the melts or partially molten rocks.
The question of the geodynamic site of TTG formation is difficult to answer solely on
geochemical grounds; indeed, all environments discussed above allow metabasalts to melt
within the garnet stability field and therefore generate sodic felsic melts, similar to TTGs.
The differences between these environments will be subtle, at best, and any interpretation
in terms of geodynamic environment requires a sound discussion of petrogenetic processes,
and a good understanding of the details of the mechanisms affecting TTG melt geochemistry.
5.6-6. PARTIAL MELTING OF AMPHIBOLITES AND CONTROLS ON THE MELT
GEOCHEMISTRY
In this section, we focus on the origin of the dominant, “true TTG” lithologies. As
demonstrated above, they belong to two main types: a high-Sr, trondhjemitic sub-series,
and a low-Sr, tonalitic to trondhjemitic sub-series.
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-6. Partial Melting of Amphibolites and Controls on the Melt Geochemistry
41
The composition of TTG rocks is a result of several different parameters. Each of them
is discussed below, in order to try and assess whether it can account for the difference
between the two sub-series.
5.6-6.1. Composition of the Source
Amphibolites (or metabasic rocks in general) actually encompass a diversity of compositions, and experimental studies have used widely different source materials (Moyen and
Stevens, 2006), from plagioclase-dominated to amphibole-dominated sources. However,
compiling experimental work shows that for major elements, the composition of the source
only marginally affects the composition of the melt. This is not surprising, considering the
generally eutectic (or at least eutectoid) nature of partial melting of Earth’s rock. Whatever
the source composition (within reasonable limits), the melting reactions and stoichiometry
will be essentially the same, yielding very similar magmas. This, of course, is not true for
trace elements, whose content in the melt is strongly tied to the source characteristics.
Interestingly, during melting of biotite and K-feldspar free lithologies (i.e., most
Archean crustal lithologies!), potassium behaves as a trace element, as there is no mineral phase that it enters other than by substituting for other ions. The K2 O contents of the
melts – and therefore, to some degree, their nature (e.g., granodioritic vs. trondhjemitic)
- therefore depends largely on the source composition (Sisson et al., 2005). High-K2 O
sources will generate high-K2 O melts, which is essentially the conclusion already arrived
at for the “high-K2 O” group (Section 5.6-5.2).
The composition of the source (or sources) of Barberton TTG rocks can be at least estimated. For a purely incompatible element (bulk repartition coefficient D = 0), the batch
melting equation (Shaw, 1970) can be simplified as C1 /C0 = 1/F (where C1 : concentration of the melt, C0 : concentration of the source and F : melt fraction). The melt fraction is,
of course, unknown. However, in experimental liquids (Moyen and Stevens, 2006), SiO2
is linearly correlated to F , such that the latter can be at least estimated. Here, we use the
following equation:
SiO2
− 1.525
F = 52
−0.011
to estimate the melt fraction.
It is therefore possible to recalculate the (possible) source composition for each sample.
The results are plotted in a multi-element, N-MORB normalized diagram (Sun and McDonough, 1989) (Fig. 5.6-12). Importantly, the concentrations predicted are only minimal
estimates, as we assumed a D value of 0: if D is higher, the source composition must consequently be higher as well. Obviously, moving to the right of the diagram (towards less
incompatible elements), the approximation becomes less correct and the source composition becomes more underestimated.
Two important conclusions can be drawn from these results:
– There are no major differences in terms of probable source compositions between the
two groups of TTGs. Both groups can derive from similar sources, suggesting that
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
42
Fig. 5.6-12. MORB-normalized (Sun and McDonough, 1989) multi-elements diagram showing the minimum trace elements concentration of
the plausible source of Barberton TTGs (see text). Same caption as Fig. 5.6-9.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
43
the differences between the low- and high-Sr groups do not reflect diversely enriched
sources. Individual plutons show an even bigger homogeneity, except the Badplaas
gneisses which display quite a large spread in calculated source composition, consistent with their composite nature in the field.
– The source of all the TTGs was an enriched MORB (>10 times chondritic for the incompatible elements). The apparent negative Nb anomaly that appears is probably an
artifact. This calculation predicts minimum estimates for the source concentrations
– the more compatible the element, the more underestimated the concentration. Owing
to the presence of phases with high affinity for Nb in the residuum (rutile), it is likely
that the D value for Nb will be rather high, and much higher than for the neighboring
elements. The predicted composition, more enriched than a Phanerozoic MORB, is however, in good agreement with the composition proposed for Archean MORBs (Jahn et
al., 1980; Condie, 1981; Jahn, 1994). Regionally, the basalts from the Komatii formation
(from the GEOROC database, http://georoc.mpch-mainz.gwdg.de/georoc/Start.asp) also
show similar compositions (Fig. 5.6-12), including a small positive Pb anomaly, which
is present in the modeled source composition. A similar, slightly enriched source composition is also predicted for (some of the) granitoids in the Pilbara Craton (Champion
and Smithies, this volume) and in Finland (Martin, 1994).
5.6-6.2. Conditions (Temperature and Depth) of Melting
To better constrain the melting conditions, we modeled the compositions of primary melts
from amphibolites as a function of the P-T conditions. The composition of the final rocks
would obviously be modified by further magmatic evolution (e.g., fractional crystallization), as discussed below (Section 5.6-6.3).
5.6-6.2.1. Principle of the model
Based on a parameterization of published experimental data, we proposed a generalized
model for vapour-absent partial melting of tholeiitic amphibolites (Moyen and Stevens,
2006). A vapour-absent scenario is favoured for reasons detailed in the cited paper, one of
the most compelling being the unlikehood of free water surviving in the crust at 10–20 kbar.
Major elements in the melt are interpolated from published melt compositions, with a linear
equation of the form (Cmelt /Csource ) = aF +b, where F is the melt fraction and a and b are
two empirically determined coefficients. The a and b coefficients used are slightly modified
from (Moyen and Stevens, 2006), the largest modification affecting the parameters for
Na2 O (we now use a = 0.025 and b = 0.60 in the garnet-amphibolitic domain; a = 0.060
and b = 0.6 in the eclogitic domain). For high melt fractions (F > 0.4), the validity of the
approximation becomes doubtful, and we simply calculate the high-F melts as a weighted
average of a F = 0.4 melt and the source. This approximation is still questionable, but not
that important, as F = 0.4 corresponds to melts with 62% SiO2 , which is less than most
of the rocks studied here.
Trace elements are calculated using an equilibrium melting equation, Kd values from
Bédard (2005, 2006), and mineral proportions interpolated from experimental data (Moyen
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
44
and Stevens, 2006). According to the conclusions above (Section 5.6-6.1 and Fig. 5.6-12),
a relatively enriched source composition is used (Sr = 240 ppm and Y = 20 ppm, within
the range of the compositions of the non-komatiitic basalts of the Onverwacht Group in
GEOROC database).
5.6-6.2.2. Variations in P-T space
The single most important parameter controlling the geochemistry of melts from metabasites is the degree of melting: higher degrees of melting (corresponding to higher temperatures) correspond to more mafic melts. Assuming both are primary melts of similar sources,
trondhjemites corresponds to melt fractions lower than ca. 20% (Moyen and Stevens,
2006), whereas tonalites reflect melt fractions up to 40–50%. Experimentally, melt fractions sufficiently high to generate a ∼65% SiO2 liquid (equivalent to the tonalites) are
attained at ca. 1000 ◦ C, below 15 kbar, but require higher temperatures as pressure goes
up (to ca. 1200 ◦ C at 30 kbar) (Moyen and Stevens, 2006). Likewise, CaO/Na2 O values
between 0.5 and 1, typical of the tonalitic rocks, correspond to the same P-T range. In contrast, the high silica, low CaO/Na2 O trondhjemites are generated at temperatures below
1000 ◦ C.
The depth of melting controls the nature of the solid phases (residuum) in equilibrium with the TTG melts. There is a potentially major difference between low to medium
pressure assemblages (amphibole and plagioclase stable, with garnet present but not abundant, and Ti mostly accommodated in ilmenite), and high pressure (eclogitic) assemblages
dominated by clinopyroxene and garnet, with rutile as the main titaniferous phase. To complicate further, even at sub-eclogitic pressures, amphibole and plagioclase are consumed
by the melting reactions, such that high melt fractions will coexist with amphibole- and
plagioclase-free restites that are mineralogically rutile-free eclogites (Moyen and Stevens,
2006).
Experimentally, both amphibolitic (Winther and Newton, 1991; Sen and Dunn, 1994;
Patiño-Douce and Beard, 1995; Rapp and Watson, 1995) and eclogitic (Skjerlie and PatiñoDouce, 2002; Rapp et al., 2003) residuum have been demonstrated to be in equilibrium with
TTG liquids. This is unsurprising, since both an eclogitic (clinopyroxene + garnet) and an
amphibolitic (amphibole + plagioclase) residuum have similar major elements compositions, except for Na2 O. Sodium is indeed less abundant in eclogitic assemblages, resulting
in high-pressure melts that are typically more sodic than their low-pressure counterparts for
a given melt fraction (Moyen and Stevens, 2006). But a more important effect is associated with the melt fraction formed. In P-T space, the melt abundance curves are positively
sloped, such that at high pressures the same melt fraction is approached only at higher
temperatures, as mentioned above.
Combining both parameters allows the identification of low-pressure liquids (relatively
high-melt fraction, sodium poor liquids: granodiorites and tonalites) and the high-pressure
liquids (lower melt fraction, more leucocratic and more sodic liquids: trondhjemites).
A major “dividing line” thus exists, separating tonalites (and granodiorites) from trondhjemites (Fig. 5.6-13). The same division is observed in Barberton TTGs, where the
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
28
35
37
46
29
36
38
45
30
37
39
31
38
40
35
36
37
38
39
40
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
32
39
41
41
33
40
42
34
41
43
“low-Sr” group plots in the tonalite and granodiorite field in O’Connor (1965) diagrams,
while the high-Sr rocks are almost exclusively trondhjemitic.
Trace elements provide slightly different information and are far more sensitive to
the pressure of melting. Indeed, trace elements will be partitioned in markedly different
ways in eclogitic (garnet-clinopyroxene-rutile) and amphibolitic (amphibole-plagioclaseilmenite) assemblages. In addition, the mode of each mineral also changes with pressure
(garnet becomes more abundant at higher pressure). Even within the realm of amphibolitic
or eclogitic residues, melt composition vary significantly as a function of depth (Moyen
and Stevens, 2006).
For the elements used here (La, Yb, Sr, Y), the main control is exerted by the abundance of high Kd phases; i.e., garnet (for Y and Yb) and plagioclase for Sr. Therefore,
trace elements in this case mostly record “pressure” information, with low pressure melts
coexisting with plagioclase but not garnet, and having low Sr but high Y and Yb contents,
42
Fig. 5.6-13. Melt composition in PT space, from parameterization of experimental data (Moyen and
Stevens, 2006); a “ThB” source (tholeiitic basalt) has been used. (a) Nature of the liquid formed (in
O’Connor (1965) systematics) as a function of the P-T conditions of melting. The thick grey line
represents 10, 30 and 50% melt (F value). Fine lines correspond to the solidus and to the mineral
stability limits (plag: plagioclase, amp: amphibole, gt: garnet). The two arrows labeled low and high
pressure melting graphically display two possible geotherm leading to the formation of trondhjemites
in one case, and granodiorites to tonalites in the second case. (b) Major element composition of the
melts. The lines correspond to iso-values of SiO2 contents and CaO/Na2 O ratios of the melts. The
thick dotted line is the “tonalite-trondhjemite divide” (panel (a), see text). (c) Sr contents of the melts
in P-T space. (d) Sr/Y values of the melts in the P-T space.
43
42
43
1
2
3
4
5
6
7
8
9
10
11
12
47
whereas at high pressures, Sr is released because of plagioclase breakdown, but Y and Yb
are locked in the garnet. Collectively, low Yb and high Sr/Y melts are produced only at
relatively high pressures (>15–20 kbar); below this threshold, higher Yb and lower Sr/Y
values are observed. Fig. 5.6-14 summarizes the geochemical trends predicted by both lowand high-pressure melting.
Combining these observations allows the clear discrimination of the two sub-series.
High-Sr melts are only trondhjemitic, and they form at high pressure (to the left of the
dividing line), plotting in the P-T space from 1000 ◦ C at 15 kbar and below to 1200 ◦ C at
30 kbar. The low-Sr group contains tonalites and granodiorites (in O’Connor’s terminology,
even if they are trondhjemites on the basis of their field appearance and mineralogy) and
forms on the high-temperature side of this divide, at pressures below 15–20 kbar.
It is worth noting that both types denote very contrasting geothermal gradients. High-Sr
TTGs formed at relatively low temperatures (probably around 1000 ◦ C), but high pressures
(>15 kbar), corresponding to a 15–20 ◦ C/km apparent geotherm. In contrast, the low-Sr
group formed at lower pressures (10–15 kbar) and comparable or higher temperatures,
corresponding to a distinct geotherm of 30–35 ◦ C/km.
The model used here is dependent on the exact parameters used (position of the mineral
stability lines, source composition, etc.). A more detailed treatment of the different cases is
presented elsewhere (Moyen and Stevens, 2006). Importantly, however, even if the actual
values are dependent on the model parameter, the same logic and the same opposition (low
P, low Sr/Y, high F melts vs. high P, low F, high Sr/Y melts) remains.
Interestingly, all of the Barberton TTGs are high-Al (Barker and Arth, 1976), and correspond to the Pilbara high-Al group of Champion and Smithies (this volume). These authors
proposed that the difference between low Al (and low Sr) and high-Al groups reflects the
depth of melting and stability of plagioclase in the residuum. In this model, both sub-series
form at pressures above the plagioclase stability field (Fig. 5.6-13), yet the geochemistry of
the melts evolves with depth, allowing a distinction between the two sub-series described
here.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Fig. 5.6-14. Modelled melting and fractionation trends in binary or ternary diagrams. The field of low-Sr (tonalites and trondhjemites), and
high-Sr (trondhjemites) are shown for comparison. Heavy arrows: melting trend from the solidus to ca. 1200 ◦ C. Grey: low pressure (13 kbar)
melting; black: high-pressure (21 kbar) melting. Thin arrows: fractionation vectors; the length of the arrow corresponds to the biggest possible
degree of fractionation (see text and Table 5.6-3). The dotted arrows correspond to models I and III, which do not fit the data. Note how
the compositional spread of each individual rock unit is “shaped” by fractionation vectors (model II, Hornblende + plagioclase most likely),
whereas their position in the diagrams is better explained by the melting trend.
42
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
25
26
27
28
29
30
31
32
33
34
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
35
36
37
38
39
40
41
42
43
43
5.6-6.3. The Role of Fractional Crystallization Following Melting
While fractionation has always been recognized as one possible process affecting TTG
composition (e.g., Martin, 1987), it is generally regarded as a minor process that only marginally affects TTG composition. However, it has recently be suggested (Bédard, 2006)
that it plays a far bigger role in shaping the trace element composition of Archean TTGs
in general (and their high Sr/Y ratio in particular), and that equivalents of Barberton trondhjemites can be generated by fractional crystallization and differentiation of tonalites. The
question is actually two-fold: (i) can fractional crystallization turn the low-Sr tonalites into
low-Sr trondhjemites; and (ii) can fractional crystallization differentiate (low-Sr) tonalites
into high-Sr tonalites?
To investigate the potential effects of fractional crystallization, we modeled the differentiation of a ca. 65% SiO2 tonalite (Table 5.6-3), using three different mineral assemblages: amphibole + biotite (model 1; Bédard, 2006); plagioclase + amphibole (model 2;
48
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
20
18
14
12
10
9
8
La
15.0
6
5
2
1
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
42
41
40
39
38
37
36
35
34
33
32
31
30
29
26
26
23
22
23
21
20
19
20
18
17
16
15
14
16
13
12
13
11
10
9
9
8
7
6
5
4
6
3
2
3
1
50
La/Yb
24
Yb
25
La
2.57
27
Sr/Y
1.42
28
Y
2.36
19.9
21.2
22.6
24.3
26.7
28.3
0.43
0.7
0.7
0.6
0.6
0.5
0.5
120.4
Sr
14.7
14.4
14.0
13.7
13.3
12.9
0.1
1.70
45.8
50.8
56.7
63.8
72.2
82.5
7.7
Na2 O
12.6
11.7
10.8
10.0
9.1
8.3
923.4
10.65
576.7
594.8
614.6
636.2
660.2
686.7
1.5
CaO
5.35
5.56
5.78
6.04
6.33
6.66
1406.0
11.75
3.78
3.39
2.97
2.49
1.94
1.32
19.04
MgO
2.00
1.46
0.86
0.18
<0
<0
15.78
2.32
1.58
0.74
<0
<0
<0
<0
FeO
<0
La/Yb
29
19
19.2
19.6
20.1
20.6
21.2
21.8
24.7
Yb
0.76
37.0
La
0.34
1.2
0.8
0.8
0.8
0.8
0.9
0.9
0.9
Sr/Y
43.6
15.5
16.1
16.7
17.4
18.1
19.0
22.3
Y
1.05
4.09
0.3
35.5
30.1
25.3
21.0
17.3
14.0
6.7
Sr
12.4
13.5
13.4
13.4
13.3
13.3
13.3
13.1
5.50
3.9
477.8
404.3
338.8
280.8
230.0
185.8
88.1
Na2 O
3.85
5.15
5.13
5.11
5.09
5.06
5.03
4.90
8.36
3.90
3.65
3.37
3.06
2.71
2.30
0.65
CaO
<0
4.77
2.37
2.24
2.09
1.92
1.73
1.51
0.63
MgO
<0
6.92
<0
2.79
2.56
2.30
2.02
1.69
1.32
<0
FeO
Model 2 (Martin, 1987 – Plagioclase + Amphibole)
SiO2
Al2 O3
Bulk repartition coefficient D
Cumulate
40.76
14.57
Fractionated liquids
5%
66.42
15.62
10%
67.85
15.68
15%
69.44
15.75
15.82
20%
71.24
25%
73.27
15.90
30%
75.59
16.00
...
80%
>100
19.58
Model 1 (Bedárd, 2006 – Amphibole + Biotite)
43
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
La/Yb
18.8
Yb
1.79
0.635
23.2
0.13
0.094
0.11
0.018
3.02
490
2.96
1.59
13
3
Yb
0.8
La
0.319
0.028
0.028
0.015
0.358
0.02
0.015
4.73
26.6
2.05
1005
12
7
Sr/Y
41.5
9.18
17.5
Y
2.47
0.603
14.1
0.037
0.138
0.07
0.018
5.42
80
11
Y
13.5
KD
Sr
0.389
0.032
0.019
0.0022
6.65
0.1
0.022
2.68
20
4.3
10.3
1.4
15
23.3
52.0
Na2 O
5.17
16
Na2 O
2.0
3.5
0.1
20
CaO
12.2
15.7
7.6
19
MgO
12.0
9.9
5.3
17
CaO
4.12
27.2
8.0
0.4
21
MgO
2.49
2
6.0
0.0
23
FeO
2.9973
Sr
560.0
0.1
12.7
24
Al2 O3
15.57
22
24.6
20.2
4.1
FeO
15.5
9.1
25.1
50.0
0.2
16.7
100.0
0.1
Table 5.6-3. Modelling fractional crystallization of a tonalite
Mineral compositions and KD ’s
60.7
36.3
30.9
AI2 O3
13.5
8.7
20.9
30.0
32.0
36.9
Major elements composition
SiO2
42.2
52.2
38.5
Amphibole
Clinopyroxene
Garnet
Ilmenite
Plagioclase
Bioite
Magnetite
Titanite
Zircon
Epidote
Allanite
Apatite
SiO2
65.14
Source (undifferentiated liquid – low SiO2 tonalite at 64–66% SiO2 )
Co
SiO2
Al2 O3
Cumulate
52.53
19.88
5%
65.80
15.34
10%
66.54
15.09
15%
67.36
14.81
20%
68.29
14.49
25%
69.34
14.13
30%
70.54
13.72
45%
75.46
12.04
...
>100
<0
80%
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
19
18
17
16
15
13
14
13
12
11
10
9
8
7
6
4
5
4
3
2
1
49
43
42
41
40
39
38
37
36
35
34
33
32
31
30
28
27
25
24
22
21
18
17
15
14
12
11
10
8
7
5
4
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
31
29
28
27
26
28
25
24
25
23
Sr
21
20
0.03
15
11
10
11
9
8
7
6
8
5
4
3
4
2
1
42
41
40
39
38
37
36
35
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
52
La/Yb
23
Yb
13.08
4.9E+09
34.8
66.7
132.7
275.4
598.9
1374.6
24
0.4
0.2
0.1
0.1
0.0
0.0
25
La
1.04
2.9E−09
26
15.0
14.9
14.9
14.9
14.8
14.8
27
Sr/Y
14.1
28
63.1
98.3
157.0
258.0
437.7
770.2
29
Y
9.20
2.2E+07
30
8.9
5.7
3.6
2.2
1.3
0.7
31
1.01
2.5E−05
32
559.7
559.4
559.1
558.8
558.5
558.1
Sr
551.5
0.07
5.44
5.74
6.07
6.45
6.87
7.36
Na2 O
25.57
15.46
3.52
2.86
2.12
1.29
0.34
<0
CaO
<0
2.64
2.48
2.47
2.46
2.45
2.44
2.43
MgO
1.89
14.63
2.39
1.70
0.94
0.09
<0
<0
FeO
<0
33
SiO2
Al2 O3
Cumulate
37.70
25.90
5%
66.58
15.03
10%
68.19
14.42
15%
69.98
13.75
20%
72.00
12.99
25%
74.29
12.13
30%
76.90
11.14
...
80%
>100
<0
Model 4 (Garnet + Epidote)
43
La/Yb
12
Yb
11.92
34.5
65.6
129.5
266.2
573.4
1302.3
13
La
0.03
0.5
0.3
0.1
0.1
0.0
0.0
3.8E+09
15
Sr/Y
15.8
16.6
17.6
18.6
19.8
21.2
1.9E−08
14
Y
7.35
60.4
89.8
136.4
212.7
341.3
565.8
71.7
16
9.7
6.9
4.8
3.3
2.2
1.4
5.5E+06
17
4.9E−04
18
1.82
Na2 O
22
588.7
620.6
656.1
696.0
741.2
792.8
11.62
5.35
5.54
5.76
6.01
6.29
6.61
2687.4
CaO
3.73
3.29
2.80
2.25
1.62
0.91
18.57
7.61
2.22
1.92
1.59
1.21
0.78
0.30
<0
MgO
2.25
1.43
0.50
<0
<0
<0
<0
17.13
<0
FeO
Model 3 (Garnet + Clinopyroxene)
SiO2
Al2 O3
Cumulate
45.35
14.80
5%
66.18
15.61
10%
67.34
15.66
15%
68.63
15.71
20%
70.09
15.76
25%
71.74
15.83
30%
73.62
15.90
...
80%
>100
18.65
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
34
dpg15 v.2007/05/23 Prn:1/06/2007; 14:54
Major elements composition are calculated using mass balance, and trace elements using Rayleigh’s law. The source composition C0 is taken as the average of
the low-SiO2 tonalites between 64 and 66% SiO2 . Partition coefficients (Kd ) are taken from Moyen and Stevens (2006). Mineral compositions are either real
minerals from TTG gneisses (Martin, 1987), or mineral in equilibrium with melts in experiments (Zamora, 2000). Three models are calculated with different
mineral proportions: model 1 (Bédard, 2006): 82% amphibole, 15% biotite, 0.5% magnetite, 0.3% titanite, 0.2% zircon, 1.5% epidote, 0.1% allanite, 0.4% apatite.
Model 2 (Martin, 1987): 39.25% amphibole, 1.5% ilmenite, 59.25% plagioclase. Model 3: 50% clinopyroxene, 50% garnet. For each model, the bulk reparation
coefficient D and the major elements cumulate composition is given; the major and trace elements composition of the fractionated liquids is given for increasing
degrees of fractionation. Impossible values for major elements (<0, meaning that the fractionation cannot process to that stage) are indicated in the left-hand side
table; corresponding trace elements values are italicized.
43
42
41
40
39
38
37
36
35
34
33
32
30
29
27
26
24
23
22
21
19
20
19
18
17
16
14
13
12
10
9
7
6
5
3
2
1
51
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
9
11
10
8
7
6
5
4
3
2
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
53
Martin, 1987); garnet + clinopyroxene (model 3) and garnet + epidote (model 4, representing high-pressure fractionation; Schmidt, 1993; Schmidt and Thompson, 1996).
In both sub-series, Al2 O3 (Fig. 5.6-5) is negatively correlated with SiO2 . This behaviour is not predicted by models 1 and 3; only models 2 (plagioclase + amphibole) and 4
(epidote + garnet) correctly predicts a decrease of Al2 O3 with differentiation. Sr decreases
with increasing SiO2 (Fig. 5.6-9(a)), as correctly predicted only by model 2. Model 4 also
predicts an uncommon behavior for Ni, which, owing to the low Kd of this element in
epidote (0.1: Bédard, 2006) and its moderate Kd in garnet (∼1.2), remains at constant concentrations or even increases. This results in the dramatic increase in Ni/Cr ratios predicted
during differentiation in model 4. Such behavior is not observed in Barberton TTGs, nor
in TTGs elsewhere in the world.
To achieve significant changes in trace elements signatures, high degrees of fractionation are required. This seems difficult to achieve, especially in high viscosity felsic melts.
Such a degree of fractionation is also difficult to achieve on geochemical grounds, as fractionation of amphibole+plagioclase (Martin, 1987), for instance, would run out of MgO
after about 40% of the crystals are removed from the melt; fractionation of biotite + amphibole (Bédard, 2006) would use up all FeO even faster, after about 20% fractionation
(Table 5.6-3). K2 O, and to a lesser degree Na2 O, would likewise be limiting factors. This
put an upper boundary on the amount of crystals that can be formed out of the melt in such
models and, accordingly, to the effect of fractional crystallization on trace elements.
Starting with a liquid with a Sr/Y of ca. 40, possible fractionation (in terms of major
elements) is sufficient to evolve a tonalite (ca. 65% SiO2 ) into a trondhjemite (ca. 72%
SiO2 ), but can not raise the Sr/Y values of the differentiated liquids above 60, 150 and 250
(models 1, 3 and 4, respectively) and Sr/Y actually decreases slightly in model 2. Only the
high-pressure fractionation models (3 and 4) have the potential to bring the Sr/Y ratios to
the high values featured by the high-Sr trondhjemites.
In summary, only models 2 (plagioclase + amphibole) and 4 (garnet + epidote) can partially fit the data. Model 2 is able to reproduce the trends observed within each rock type,
but has only a limited effect and can barely fractionate the tonalites into trondhjemites. It is
also unable to change low-Sr rocks into high-Sr rocks and can also not account for the high
Sr/Y values in the (high-Sr) trondhjemites, as the fractionation of amphibole + plagioclase
has no noticeable effect on Sr/Y values of the melts. Model 4, on the other hand, has a
more pronounced effect on the melt compositions, and could result in evolution of low-Sr
tonalites into high-Sr trondhjemites. But the fit with the data is poorer (elements such as Sr
and Ni are not convincingly modeled). Furthermore, model 4 calls for fractionation of garnet and epidote, a high pressure (>20 kbar: Schmidt, 1993; Schmidt and Thompson, 1996)
and high water activity assemblage, regardless of whether the high Sr/Y is related to high
pressure melting (as proposed Section 5.6-6.2.2), or to high-pressure fractionation. Nevertheless, it points to evolution at pressures >20 kbar for the high-Sr sub-series, but such
pressures are not required for the low-Sr group. Finally, while fractionation of epidote +
garnet can change a low-Sr liquid into a high-Sr liquid, the reverse is not true and it appears
impossible to fractionate a low-Sr tonalite formed at shallow depth (see Section 5.6-6.2.2)
under high-pressure conditions!
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
54
Collectively, it seems that if fractionation played a role in the geochemical evolution of
Barberton TTGs, it was only minor. The geochemical trends for at least some of the plutons
are shaped by late fractionation (amphibole + plagioclase), probably reflecting liquidcrystal separation during emplacement. It is also possible that the high-Sr (and high Sr/Y)
signature of some deeply generated trondhjemites was enhanced by some high-pressure
fractionation. But fractionation cannot account for the difference between the low- and
high-Sr sub-series: they represent two fundamentally different sub-series, reflecting different conditions of melting (Fig. 5.6-14). In addition, fractionation can barely explain the
difference between tonalites and trondhjemites, and in all likelihood this difference also
reflects different melting conditions (temperatures).
5.6-6.4. Possible Interactions with the Mantle
If melting occurs at great depth (whatever the context, see below), it will most likely occur
below a peridotite layer. Therefore, the TTG magma rising to the surface will have to cross
a large volume of peridotite and will most likely interact with it, resulting in the formation
of “hybrid” TTGs (Rapp et al., 2000; Rapp, 2003; Martin et al., 2005). It has been proposed
(Smithies, 2000; Martin and Moyen, 2002) that the secular increase of Mg#, Ni and Cr in
TTGs reflects progressively deeper melting, allowing more pronounced interactions with
the mantle. At the extreme end of this spectrum of melt-mantle interactions is the formation
of “sanukitoids” (Martin et al., 2005). Sanukitoids are characterized by both elevated Mg,
Ni and Cr contents and significant LILE and REE enrichments, typically with relatively
high K/Na ratios (Moyen et al., 2003). High HFSE levels are also common. This association is not found in any of the Barberton TTG, and we see no evidence for interactions
between TTG melts and the mantle in the BGGT.
5.6-7. SUMMARY AND GEODYNAMIC IMPLICATIONS
5.6-7.1. Petrogenetic Processes for Individual Plutons
5.6-7.1.1. The ca. 3.55–3.50 Ga Steynsdorp pluton
Despite only relatively few analyses being available, the Steynsdorp pluton appears to be
made up of two components; low-Sr tonalites and high-K2 O granodiorites. The existing
data and the discussion above suggest that the tonalitic component represents relatively low
depth, high melt fraction liquids from amphibolites. The granodiorites, interleaved with the
tonalites (Kröner et al., 1996), display the characteristics trends and high-K2 O nature of the
“secondary” liquids, which formed by remelting of pre-existing TTG, probably equivalents
of the associated tonalites.
5.6-7.1.2. The ca. 3.45 Ga group (Stolzburg and Theesprui plutons)
Intrusive phases of the Stolzburg and Theespruit plutons are fairly homogeneous. They
are leucocratic trondhjemites, mostly belonging to the high-Sr, high-pressure, low-melt
fraction group. Isotopically, their source was the most depleted of the studied rocks.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-7. Summary and Geodynamic Implications
55
A somewhat surprising feature of the ca. 3.45 Ga high-Sr TTGs, however, is that despite
their probable deep origin (>20 kbar; e.g., more than 60 km), there is no clear evidence for
interaction with the mantle in the geochemical signature of this group.
5.6-7.1.3. The 3.29–3.24 Ga Badplaas gneisses
The Badplaas gneisses are the most complex and composite unit of the BGGT plutons.
They include all 4 rock types identified regionally: high- and low-Sr “true” TTGs, together with high-K2 O rocks and matching “melt-depleted” samples, both probably related
to re-melting of the newly emplaced TTGs. The true TTGs belong to the two sub-series,
demonstrating that the Badplaas gneisses were formed from sources at different depths.
Therefore, it seems that the Badplaas gneisses recorded a long (ca. 50 My) and complex
history of melting of a vertically extensive source region, accretion of a “proto Badplaas
terrane” and remelting of this terrane, possibly during the ca. 3.22 Ga subduction-collision
event. Proper interpretation of the geochemistry of the Badplaas gneisses, however, would
require a more detailed, field-constrained study of the different units, which is beyond the
scope of the present work.
5.6-7.1.4. The ca. 3.23–3.21 Ga Nelshoogte pluton
The Nelshoogte pluton is a composite intrusion, made up of early trondhjemitic phases
belonging to the low-Sr group (although quite close to the boundary with the high-Sr
sub-series), intruded by a later set of low-Sr tonalites, clearly cutting across the earlier
lithologies. This indicates a succession of melting conditions at moderate depths but with
increasing temperatures, consistent with the emplacement of this pluton during orogenic
collapse of the BGGT 3.22 Ga “orogen” (Belcher et al., 2005). The relatively enriched isotopic characteristics of the Nelshoogte pluton are consistent with melting of the preexisting
Onverwacht (or even Fig Tree) supracrustals, and also support this model.
5.6-7.1.5. The ca. 3.23–3.22 Ga Kaap Valley tonalite
The Kaap Valley pluton is almost exclusively made of phases belonging to the low-Sr
group, pointing to shallow, high-melt fraction melting. Isotopic characteristics also suggest
a slightly enriched (Onverwacht-like) source, whereas the emplacement history is also consistent with syn-exhumation intrusion. The relatively high REE contents of the Kaap Valley
pluton (compared to the other TTGs) has been interpreted as precluding simple derivation
by melting of a common source (Robb et al., 1986). Indeed, the isotopic data also points to
a slightly different origin for the Kaap Valley tonalite compared to the other TTG plutons.
We suggest that these differences mostly reflect melting of the (relatively enriched) Onverwacht supracrustals (mostly mafic and ultramafic lavas, but possibly with incorporation
of a minor sedimentary component). The unique nature of the Kaap Valley pluton would,
therefore, reflect both a slightly different (more fertile) source and a higher temperature of
melting compared to the other TTG plutons, the combination of both parameters resulting
in higher melt fractions and the generation of a dominantly tonalitic pluton, unique in the
BGGT. It seems, therefore, that the Kaap Valley pluton mostly reflects partial melting of
the base of an already formed crust (Onverwacht Group-like).
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
42
56
43
5.6-7. Summary and Geodynamic Implications
57
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
1
40
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
15
14
13
12
11
10
9
8
7
6
5
4
3
2
16
41
43
Fig. 5.6-15. Geodynamic model for the evolution of the BGGT, with emphasis on the formation and
emplacement of TTG plutons. On the right hand side, a time scale shows the position of the cartoons
in the global evolution of the BGGT. Note that, for the cartoons on the left, the time scale is distorted.
Also note that the scale is not a stratigraphic scale, as the younger stages are at the bottom. Left-hand
side cartoons are approximately at the same scale, looking towards the (present-day) northeast; the
front section of each block corresponds to a NW-SE cross-section. In each cartoon, the active plutonism is in black, whereas rocks that have already been emplaced are in grey. Symbols denote the
melting zone: stars are for melting of amphibolites (grey: deep, generating high-silica trondhjemites,
black forming low-silica tonalites); white triangles denote melting of already formed felsic crust; inverted black triangles are for the melting of the mantle. In the top stage (Steynsdorp), two alternatives
are proposed: intra-plate accretion of an oceanic plateau, followed by remelting at its base generating low-pressure TTGs (left), or low-pressure melting at the base of a tectonic stack of oceanic crust.
The last cartoon shows more or less the relative positions of individual geological elements (that have
been only marginally modified by the later, ca. 3.1 Ga events). Plutons: B: Badplaas, N: Nelshoogte,
KV: Kaap Valley, S: Stolzburg, Ts: Theespruit. Structures: IF: Inyoka Fault, ISZ: Inyoni Shear Zone.
Cartoons are modified from Moyen (2006), the top three are inspired from Lowe (1999).
17
42
5.6-7.2.1. Ca. 3.23–3.21 Ga: main event of terrane accretion
The dominant geological event that shaped the present-day structure of the belt occurred at ca. 3.23 Ga. Structural (de Wit et al., 1992; de Ronde and de Wit, 1994;
de Ronde and Kamo, 2000; Kisters et al., 2003) as well as metamorphic (Dziggel et
al., 2002; Stevens et al., 2002; Kisters et al., 2003; Diener et al., 2005; Dziggel et al.,
2005; Diener et al., 2006; Moyen et al., 2006) studies suggest collision (or arc accretion) between two relatively rigid blocks, separated by the Inyoni–Inyoka tectonic system (Lowe, 1994). The western terrane has largely been overprinted by the ca. 3.25–
3.21 Ga rocks (Fig Tree lavas and TTGs), but was probably built on a nucleus of
slightly older (3.3–3.25 Ga: de Ronde and de Wit, 1994; Lowe, 1994; Lowe and Byerly, 1999; Lowe et al., 1999; de Ronde and Kamo, 2000) mafic and ultramafic lavas,
possibly an oceanic plateau of some sort. The eastern terrane is better preserved and
was at this time a composite unit including old lavas and sediments intruded by ca.
3.45 Ga TTGs and overlain by still younger mafic/ultramafic lavas. It is interpreted to
represent an oceanic plateau that was modified by a relatively minor subduction event
(see below). The accretion itself occurred via under-thrusting (subduction?), and the
eastern, high-grade Stolzburg terrane probably represents the lower plate of this event
(Fig. 5.6-15).
In addition to the geochemical information presented above, the geodynamic evolution
of the Barberton Greenstone Belt has been discussed in other papers in this volume. Our
geochemical and geodynamical conclusions fit with this model, and allow us to refine it
in some aspects. As the geological history of each event is partially erased by subsequent
events, it will be presented backwards, starting with the youngest.
5.6-7.2. Geodynamic Model
43
42
43
58
In this context, the ca. 3.29–3.21 Ga plutons record the transition from pre-collision to
post-collision magmatism. The earliest phases formed by deep melting (high Sr parts of
the Badplaas gneisses) and their ages correspond to the accretion stage of the BGGT, most
probably in a magmatic arc (de Ronde and Kamo, 2000; Kisters et al., 2006). The latest
phases (low-Sr rocks in the three units) formed by relatively shallow (10–12 kbar) melting of amphibolites, possibly parts of the Onverwacht Group. The transition from low-Sr
trondhjemites (bulk of the Nelshoogte pluton, part of Badplaas gneisses) to low-Sr tonalites
(late phases in the Nelshoogte pluton, Kaap Valley pluton) reflect increasing temperatures
at the base of the collapsing pile, as commonly observed in post-orogenic collapse (Kisters
et al., 2003). Some of the early rocks underwent intracrustal remelting, more or less at the
same time (mostly in the Badplaas pluton). Field and structural studies demonstrate that at
least some of these plutons formed during orogen parallel extension, all of which is consistent with lower crustal melting of the thickened, dominantly mafic crust during orogenic
collapse, and/or possibly during slab breakoff.
From south to north (i.e., from Badplaas to Kaap Valley), there is an overall evolution towards younger and lower silica rocks, reflecting the switch from syn-subduction or
collision to syn-collapse magmatism: the latest, collapse-related magmatic event is better
represented in the northern plutons. This could reflect some along-strike differences between the southern segment of the orogen, which involved an already rigid continental
nucleus (the already-formed Stolzburg terrane), and the northern segment, where no evidence for rigid crust is documented and which could have been a less consolidated volcanic
arc at the time.
5.6-7.2.2. Ca. 3.45 Ga: accretion of the Stolzburg domain
The origin of the continental Stolzburg domain is somewhat obscured by the dominant,
ca. 3.23–3.21 Ga collision. The composition, mirroring a deep source, of the 3.45 Ga
old Stolzburg and Theespruit plutons suggests that they could have intruded as suprasubduction zone plutons into a small, mafic to ultramafic crustal block. This is consistent
with their shallow level of emplacement. Existence of a still older crust (the lower Onverwacht Group and the Steynsdorp pluton) suggests that this subduction occurred along
the margin of a pre-existing “proto-continent” (whatever its nature was, the abundance of
komatiites suggests that it was probably an oceanic plateau). After the emplacement of the
TTG plutons, renewed komatiitic volcanism at ca. 3.45–3.40 Ga has been interpreted as
reflecting the rifting of the newly formed crustal nucleus (Lowe et al., 1999).
5.6-7.2.3. Ca. 3.55–3.50 Ga: the early Steynsdorp continental nucleus
The ca. 3.55–3.50 Ga TTGs of the Steynsdorp pluton apparently formed by shallow melting of amphibolite (and quick remelting of the newly formed felsic lithologies). We suggest
that this could represent the very start of the cratonization process, through remelting of
the lower part of a thick pile of mafic rocks.
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-8. Discussion
5.6-8. DISCUSSION
5.6-8.1. The Different Sub-Series of TTGs
59
An important result of this work is the identification of three main types of magmas, all belonging to the wide group collectively referred to as “TTGs”, and in fact all being high-Al
TTGs (Barker and Arth, 1976; Champion and Smithies, this volume). First is a group of
relatively potassic rocks, mostly granites and granodiorites, with some trondhjemites, is derived from melting of relatively felsic, enriched sources such as pre-existing TTGs or felsic
components (sediments, felsic lavas) of the supracrustal pile. The “true” TTGs are themselves differentiated into high-Sr TTGs that are mostly trondhjemites, and low-Sr TTGs
ranging from tonalites to trondhjemites and granodiorites. While fractional crystallization,
probably late and emplacement-related, does play a role in shaping the geochemical trends
of individual plutons, it cannot explain the first-order differences between low- and highSr sub-series, and between tonalites and trondhjemites: the sub-series identified herein
correspond to differences between primary melts. The high-Sr sub-series formed by high
pressure (>20 kbar) melting of amphibolites, whereas the low-Sr sub-series formed by
lower-pressure (and relatively high temperature) melting of amphibolites.
The difference between the three sub-series is important, as each of them corresponds
to a significantly different combination of sources and P-T conditions of melting. Any
geodynamic reconstitution or tectonic model based on “TTG” magmatism should take into
account these differences, as they represent important constraints on our understanding of
the crustal evolution of Archean cratons.
5.6-8.2. Comparison with the Pilbara Granitoids
5.6-8.2.1. Geochemical observations
TTG granitoids of the same period (3.5–3.2 Ga) are a major lithology in the Pilbara Craton
(Champion and Smithies, this volume), such that it is worth drawing some comparisons. In
the Pilbara, two main suites of “TTG” (or related) rocks are found; a high-Al, high-Sr group
and a low-Al, low-Sr group. In Barberton, two groups of TTGs are also observed (low-Sr
tonalites and trondhjemites, and high-Sr trondhjemites), but both of these would fall within
the definition of high-Al, high-Sr rocks in the Pilbara. The Pilbara high-Al series ranges
from ca. 65–72% SiO2 , broadly corresponding to the range observed in Barberton (low Sr)
TTGs. However, the low-SiO2 series (<68% SiO2 ) seem to be less common in the Pilbara
than in Barberton (where they form the Kaap Valley pluton). True low-Al series rocks are
completely missing from the Barberton rock record. In the Pilbara, the two series are not
temporally or spatially distinct, with no clear logic behind the repartition of the types. In
contrast, in the BGGT there is a clear repartition of the two rock types; the older plutons
(ca. 3.45 Ga) are high-Sr trondhjemites, whereas the younger (3.23–3.21 Ga) plutons are
of the low-Sr series type. Only the complex Badplaas gneisses show, on a smaller scale
of a few kilometers, the same degree of internal complexity, both in terms of time and of
geochemistry, but again, low-Al rocks are missing from this unit.
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
60
Finally, some of the Pilbara TTG (mostly from the low-Al group) are enriched in LILE,
including K2 O, which makes them granodioritic rather than trondhjemitic (“transitional
TTGs”). They can be compared to the “high K2 O/Na2 O” rocks that we have identified in
Barberton TTGs, although we observe a compositional gap between the high-K2 O rocks
and the “ordinary” TTGs, rather than the continuous evolution recorded in the Pilbara. In
addition, most LILE-enriched TTGs from the Pilbara belong to the low-Al series, which is
missing from Barberton. The high K2 O rocks are also rarer in Barberton, where they form
minor phases of composite plutons, such as Steynsdorp or Badplaas, and (apart from some
dykes) are largely missing from the simple, monogenetic plutons like Stolzburg or even
the Nelshoogte and Kaap Valley plutons. However, a large part of what is referred to as the
“late GMS suite” (the 3.1 Ga batholiths, clearly distinguished from the older TTG magmatism in our case) in Barberton is actually, geochemically, quite similar to the transitional
TTGs of the Pilbara, including relatively low Sr and Al contents (Anhaeusser and Robb,
1983; Yearron, 2003; Belcher et al., submitted). This suggests that the classical distinction
between “TTG gneisses” and “late potassic plutons” (e.g., Moyen et al., 2003) might not be
that clear, as the same type of rock can be regarded either as “a LILE-enriched component
of the TTG gneisses”, or “late potassic plutons”, depending on the field relationships. In all
cases, the interpretation proposed is quite similar: all groups of rocks are interpreted to reflect the melting of “a LILE-enriched, ‘crustal’ component” (Champion and Smithies, this
volume),“pre-existing felsic lithologies (e.g., tonalites)” (this work), or “the ca. 3.5–3.2 Ga
TTG basement” (Anhaeusser and Robb, 1983; Belcher et al., submitted).
5.6-8.2.2. Petrogenetic models
The petrogenetic models proposed for the Pilbara (Champion and Smithies this volume)
and Barberton (this work) granitoids are quite similar. The “normal” TTGs are regarded
as the products of amphibolite melting at different depths, resulting in the distinction between low- and high-Al sub-series. The comparison between Barberton and Pilbara data
suggests that the high-Al series can further be subdivided into a low- and high-Sr group.
Superimposed on this classification, we observe in Barberton a difference in melt fractions
(SiO2 contents) that leads us to propose different geothermal gradients, as well as different
depths of melting, which is apparently not the case in the Pilbara. Fractional crystallization
and interactions with a mantle wedge are, in both cases, regarded as minor processes, at
best.
“Transitional” (LILE-enriched, high K2 O/Na2 O) facies are regarded as melts of more
felsic lithologies. In the Pilbara, Champion and Smithies (this volume) propose that this
occurs both at high and low depths, resulting in transitional TTGs belonging both to the
low- and high-Al groups. These are interpreted to form at the same time, and in the same
regions, as the other TTGs. In Barberton, they mostly belong to the low-Sr group (as in the
Pilbara); high-Sr samples are apparently associated with high-Sr “true” TTG plutons (e.g.,
Stolzburg). We propose that, rather than coeval magmas, they more commonly correspond
to later remelting of already emplaced TTGs, at mid-crustal depths.
Our geodynamical inferences differ from these proposed by Champion and Smithies. In
the Pilbara, the interlayering of all types of rocks (low and high Al, “normal” and “tran-
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5.6-9. Conclusions
61
62
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
39
38
37
36
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
26
27
28
29
30
31
32
33
34
35
36
37
38
D. Champion and H. Smithies kindly supplied an early draft of their manuscript in this
volume that was highly thought-provoking and allowed us to draw fruitful comparisons
between our two models. A detailed review by Jean Bédard greatly helped to improve
both the content and the form of the manuscript. JFM’s post-doctoral fellowship at the
University of Stellenbosch was funded by the South African National Research Foundation
(grant GUN, 2053698) and a bursary from the Department of Geology, Geography and
Environmental Sciences. Running costs were supported by a NRF grant awarded to AFMK
(grant no. NRF, 2053186). Access to lands and the hospitality of farmers and residents in
and around the town of Badplaas is greatly appreciated.
1
36
38
40
39
40
41
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
37
39
41
1
38
40
42
2
39
41
43
can be regarded as a record of Archean subduction. It seems likely that a similar distinction
between low- and high-Sr sub-series may be possible throughout the Archean: at least some
recent studies (e.g., Benn and Moyen, submitted; Champion and Smithies, this volume)
suggest that this distinction applies to other Archean cratons, as well.
The degree of enrichment of the source is also recorded to some degree in the composition of the TTGs, and we can distinguish between “normal TTGs” (melts from amphibolites) and “high-K2 O samples” (melts from more felsic lithologies – either older TTGs,
or felsic lavas/sediment components in the source). It is quite possible that further studies
will demonstrate further distinctions between more or less enriched sources.
Subordinate factors controlling the composition of TTGs include later fractional crystallization (although reasonable degrees of fractionation do not hugely modify the geochemistry of these rocks) and interaction with mantle rocks (implying some form of
lithosphere-scale imbrication of mantle and crust rocks). While minor on a craton scale,
these processes can locally be important in the petrogenesis of one specific rock unit, and
cannot be a priori ignored.
In the BGGT, the evolution from “shallow” tonalites at 3.55–3.50 Ga, to “deep” trondhjemites at 3.45 Ga, to “shallow”, complex tonalites and trondhjemites at 3.29–3.24 Ga
probably mirrors the formation and evolution of the eastern segment of the Kaapvaal Craton, from the generation of an early crustal nucleus, its subsequent growth via the addition
of new material generated along a subduction margin, to its final accretion (and reworking)
in a collisional orogen.
40
42
sitional” TTGs) leads them to propose a model of essentially intracrustal melting of a
dominantly basaltic stack, with locally more felsic layers. Progressive differentiation of
the crust would lead to increasingly felsic, and increasingly more crustal, sources and account for the relative abundance of transitional TTGs in the later stages. In contrast, in
Barberton, the time and space repartition of the different rock types allows them to be fitted into the framework of a “plate tectonics” model (at least at 3.2 Ga, possibly at ca. 3.45
Ga). Here too, the progressive “maturation” of the crust eventually results in the formation
of “potassic” magmas (corresponding to the ca. 3.1 Ga GMS suite of the BGGT), forming
well defined, younger, clearly distinct batholiths that contrast with the less well-defined
“transitional” phase of the Pilbara.
It would then appear that the two cratons followed a somewhat different early evolution.
The Pilbara Craton, from 3.45 to 3.3 Ga, apparently evolved essentially in an intra-plate
setting (oceanic plateau; see also Van Kranendonk et al. and Smithies et al., this volume), reflected by heterogeneous sources and depth of melting for the granitoids of this
time. In contrast, after the initial accretion of “shallow” TTGs (probably through intraplate
processes, as well) at ca. 3.55 Ga, the BGGT shows very homogeneous, deeply-originated
TTGs at ca. 3.45 Ga. We interpret this to be subduction related. This suggests that some
sort of arc fringed the oceanic plateau that was the proto-BGGT at ca. 3.45 Ga, in contrast
with the Pilbara nucleus, which is devoid of any such structure.
However, at ca. 3.2 Ga, the evolution of both cratons again becomes similar; e.g., a
modern-style arc setting in the Pilbara, based on the geochemistry of ca. 3.12 Ga lavas
(Smithies et al., 2007), and a collisional orogenic setting in the Barberton, based on the
geochemistry of the ca. 3.2 Ga plutons (Stevens and Moyen, this volume).
24
41
43
ACKNOWLEDGEMENTS
25
42
Far from being the homogeneous, monotonous group of rocks that they are commonly
assumed to be, TTGs are a complex, composite group encompassing a large family of plutonic rocks showing evidence for a diversity of processes, both in term of emplacement
history and geochemistry/petrogenetic history. This suggests that specific attention should
be paid to the details of the field relations and geochemistry of the TTG gneisses and to elucidate their intricate histories, as they are more than the simple “basement” to (apparently)
more interesting supracrustal lithologies.
The most significant information recorded by the TTG geochemistry is linked to the
depth of melting of the source amphibolite. Geochemistry of the high-Al TTGs of the
BGGT allows the differentiation between two “sub-series”; a high-pressure and relatively
low temperature sub-series of mostly leucocratic trondhjemites (“high Sr sub-series”), and
a lower-pressure and higher temperature sub-series of considerably more diverse rocks,
ranging from tonalites (and even diorites) to trondhjemites and granodiorites (“low Sr subseries”). The geothermal gradients of both sub-series record, together with the established
tectonic framework of the BGGT, that only the high pressure sub-series (corresponding
to the ca. 3.45 Ga plutons and part of the 3.29–3.24 Ga Badplaas gneisses in Barberton)
5.6-9. CONCLUSIONS
43
42
43
Abstract
Journal of Structural Geology 28 (2006) 1406e1421
www.elsevier.com/locate/jsg
Progressive adjustments of ascent and emplacement controls
during incremental construction of the 3.1 Ga Heerenveen
batholith, South Africa
R.W. Belcher, A.F.M. Kisters*
Received 9 January 2006; received in revised form 19 April 2006; accepted 2 May 2006
Available online 5 July 2006
The Heerenveen batholith is part of a suite of areally extensive, shallow-crustal granitoid plutons intruded during the last regional phase of
tectonism and NW-SE subhorizontal shortening recorded in the Mesoarchean Barberton granitoid-greenstone terrain of South Africa at 3.1 Ga.
Intrusive relationships allow at least four main successively emplaced intrusive stages to be distinguished. Each of these shows distinct geometries and intrusive styles that provide evidence for the progressive change of emplacement controls during the incremental construction of the
Heerenveen batholith. The earliest sheet-like granitoids intruded as foliation-parallel sills along the shallowly dipping basement gneissosity, emphasizing the role of favourably inclined pre-existing wall-rock anisotropies for granite emplacement during the early stages of pluton assembly.
Continued sheeting and coalescence of sheets provided the thermal ground preparation that led to the formation of larger, coherent magma bodies and the main phase of homogeneous, commonly megacrystic granites. These megacrystic granites form the central parts of the Heerenveen
batholith, and are interpreted to represent steady-state magma chambers. The introduction of the rheologically weaker melt bodies into the shallow crust resulted in the nucleation of conjugate synmagmatic transpressive shear zones around the central granites. The shear zones correspond
to several km-wide zones of shear zone-parallel granite sheeting. This stage marks a dramatic switch in emplacement styles. While the initial
stages of magma emplacement were largely determined by factors external to the magma, most importantly the pre-existing wall-rock anisotropies, subsequent stages are dominated by factors intrinsic to the magma, namely strain localization and partitioning along melt-bearing zones
during syntectonic plutonism. The associated melt transfer along these zones is independent of pre-existing structures and mainly related to
buoyancy- and strain-induced melt ascent. The last granites of the Heerenveen batholith are post-tectonic. They intrude as either plugs or stocks
of seemingly random orientation, but display a clear control by wall-rock anisotropies where they are in contact with the country rocks.
On a regional scale, the different phases of the Heerenveen batholith describe an overall zonation of central homogeneous granites enveloped
by composite, sheeted and sheared margins. This pattern is typical for most of the large 3.1 Ga granite batholiths in the Barberton granitoidgreenstone terrain. We suggest that the sequence of progressively changing emplacement controls and the formation of steady-state magma
chambers described here for the Heerenveen batholith may be of wider application to other zoned and/or incrementally assembled batholiths.
Ó 2006 Elsevier Ltd. All rights reserved.
of discrete melt batches that often take the form of sheet-like bodies (e.g. Ingram and Hutton, 1994; Wiebe and Collins, 1998;
Miller and Paterson, 2001; Mahan et al., 2003; Archanjo and
Fetter, 2004). Geochronological data on many of these sheeted
granites have confirmed that pluton growth is incremental and
may occur over up to several million years rather than representing short-lived single-stage events (e.g. Johnson et al., 2003;
Coleman et al., 2004; Glazner et al., 2004; Westraat et al., 2005).
Keywords: Archean granites; Barberton terrain; Sheeted granites; Incremental emplacement; Magma chamber formation
1. Introduction
Large and seemingly homogeneous granite plutons are increasingly recognized as being constructed through the assembly
* Corresponding author.
E-mail address: [email protected] (A.F.M. Kisters).
0191-8141/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsg.2006.05.001
1407
The Heerenveen batholith is exposed as an elongate, approximately 30 km long and 15 km wide, NNE-SSW orientated body along the main eastern escarpment in South
Africa that offers a topographic relief of some 600 m. The
granitoids are intrusive into older 3225e3450 Ma TTG
gneisses and supracrustal greenstones that underwent regional
deformation and mid-amphibolite facies metamorphism at ca.
3. Heerenveen batholith
size, covering in total, an area in excess of 24 000 km2
(Fig. 1). The plutons show a broad compositional range from
trondhjemites to syenites (Anhaeusser and Robb, 1983), but
the most common rock types are granodiorites, monzogranites
and syenogranites, collectively referred to as the GMS suite
(Yearron, 2003). Geochemically, the batholiths are medium
to high potassium, calc-alkaline, metaluminous to slightly
peraluminous I-type granitoids (Hunter, 1973; Anhaeusser
and Robb, 1983; Anhaeusser et al., 1983; Robb et al., 1983;
Yearron, 2003). In outcrop the batholiths have thin
(<1000 m), generally tabular geometries, and based on textural
characteristics and the low-grades of metamorphism recorded
in e.g. the adjacent Barberton greenstone belt, relatively shallow (<5 km) emplacement levels are suggested (Hunter,
1973; Anhaeusser et al., 1983; Robb et al., 1983). Internal contacts between different intrusive phases reveal an overall
sheeted architecture. Granite sheeting is particularly well
developed along the margins of the batholiths, consisting of
a multitude of compositionally distinct, subvertical and/or
shallowly dipping sheets. These up to several km-wide
sheeted margins surround a core of more massive granitoids
(Anhaeusser et al., 1983; Westraat et al., 2005; Belcher and
Kisters, 2006).
Available age data indicate that the composite GMS batholiths have intruded over a time span of approximately 15 Ma
between ca. 3100 and 3115 Ma (Kamo and Davis, 1994;
Westraat et al., 2005). Traditionally, the 3.1 Ga granitoids
have either been interpreted as having intruded into an extensional and/or transtensional setting (De Ronde and De Wit,
1994; Kamo and Davis, 1994) or forming anorogenic granitoids
emplaced after the tectonic assembly of the Barberton granitoidgreenstone terrain (e.g. Anhaeusser et al., 1983). More recent
structural work has highlighted the presence of pervasive magmatic- and solid-state fabrics, synmagmatic shear zones and the
progressive deformation of the granitoids of the GMS suite
(Westraat et al., 2005; Belcher and Kisters, 2006), first described
by Jackson and Robertson (1983 ). The fabric development, orientation and kinematics of the synmagmatic shear zones, and
folding and/or boudinage of intrusive phases all point to the emplacement of the GMS suite during regional NW-SE subhorizontal contraction (Belcher and Kisters, 2006). Both the
intrusive ages and synmagmatic deformation of the batholiths
show that the GMS suite plutonism was concurrent with the
last phase (D3; 3126e3084 Ma, De Ronde et al., 1991) of regional NW-SE shortening and associated folding and thrusting
documented in the Barberton greenstone belt (De Ronde and
De Wit, 1994; Kamo and Davis, 1994).
The formation of granites is almost invariably linked to orogenic environments and the close correlation between deformation zones and regions of melt transfer and granite
emplacement are well documented (e.g. D’Lemos et al.,
1992; Ingram and Hutton, 1994; Brown and Solar, 1998).
The presence of low-viscosity melts in a deforming crustal
section may lead to strain localization that may not only lubricate but also trigger shear zones resulting in deformation rates
in the melt-bearing zones that are substantially higher than average crustal strain rates (e.g. Davidson et al., 1994; Rutter and
Neumann, 1995; Tommasi et al., 1995; Gerbi et al., 2004).
Similarly, zones of magma transfer and granite emplacement
may also result in strain partitioning, and magmatic arcs of
e.g. transpressional orogenic belts can often be shown to localize the non-coaxial component of the bulk strain (Fitch, 1972;
Tikoff and Teyssier, 1994; De Saint-Blanquat et al., 1998;
Vigneresse and Tikoff, 1999).
Given the effects of strain partitioning and localization
around magmas in conjunction with the progressive assembly
of many granitoid plutons, one must expect progressive adjustments or transient switches of the controls and styles of emplacement in and around the sites of emplacement or zones
of melt transfer (e.g. Marsh, 1982; Furlong and Myers,
1985; Cruden, 1990; Bergantz, 1991). There are numerous
cases documenting the positive feedback effect between melting and deformation (e.g. Karlstrom et al., 1993; Ingram and
Hutton, 1994; Brown and Solar, 1998; De Saint-Blanquat
et al., 1998; Neves et al., 2000). However, examples where
earlier granite phases can be shown to modify the emplacement styles and controls of ascent of later granite batches of
composite plutons are only rarely documented. This may be
expected in large plutonic bodies where the intrusion of subsequent magma batches is likely to obscure the emplacement
controls of earlier granite phases.
The purpose of this paper is to examine and document the
systematic changes and adjustments of intrusive styles and
emplacement controls recorded by successive granitic magma
batches during the incremental construction of a large, composite batholith. This study focuses on the 3.1 Ga Heerenveen
batholith in the Barberton granitoid-greenstone terrain. The
Heerenveen batholith is composed of a number of texturally
and compositionally distinct granitic phases, comprising
more massive granitoids in the centre, surrounded by kmwide zones of heterogeneous sheeted granites. The paper starts
by describing the intrusive and relative age relationships between different granite phases, their geometries and magmatic
and solid-state fabrics. The distinct intrusive styles and fabric
developments in the different granite phases are then discussed
in terms of the progressive modification of emplacement styles
induced by earlier granite phases.
2. Regional geology
The Heerenveen batholith is one of the several areally
extensive 3.1 Ga plutons located in the Mesoarchean (3.5e
3.1 Ga) Barberton granitoid-greenstone terrain in South Africa.
Individual plutons range from several 100 to >5000 km2 in
1408
a 500e1500 m wide zone. The eastern margin of the batholith
consists of a several km-wide zone in which a multitude of mwide granitoid sheets intrude the shallow SE-dipping gneisses
in a lit-par-lit fashion (Fig. 3). This zone corresponds to the
‘‘marginal migmatite zone’’ originally mapped by Anhaeusser
et al. (1981) and Anhaeusser and Robb (1983). An exception
to this rather gradational contact occurs in the SE, where the
granitoids abut sharply against the subvertical NE-trending
Schapenburg schist belt (Anhaeusser, 1983; Stevens et al.,
2002).
Several texturally and mineralogically distinct phases are
recognized in the Heerenveen batholith, each of which has
characteristic internal geometries and fabrics (Table 1). Four
main intrusive phases can be distinguished that, based on
Fig. 1. Simplified geological map of the Mesoarchean Barberton granitoid-greenstone terrain (after Anhaeusser et al., 1981) showing the distribution of intrusives
of the 3.1 Ga granodioriteemonzoniteesyenogranite (GMS) suite. Age data for the GMS suite plutons are from Kamo and Davis (1994). Inset: location of the
Barberton granitoid-greenstone terrain on the Archean Kaapvaal Craton in southern Africa.
3225 Ma (Stevens et al., 2002). The northern and southern extents of the Heerenveen batholith are concealed by younger
cover sequences. The roof rocks of the pluton are nowhere exposed, while the floor of the pluton is locally transected along
the eastern and southeastern margins of the Heerenveen batholith, exposing shallowly dipping gneisses of the Badplaas
basement. In the west, the Heerenveen batholith is intrusive
into E-trending subvertical TTG gneisses and minor amphibolitic remnants of the older, trondhjemitic Rooihoogte basement. Within 1e2 km of this contact, the basement gneisses
are rotated into parallelism with the curvilinear western margin of the batholith (Fig. 2). This contact and the transition
from country-rock gneisses via sheeted granites and intrusive
breccias into more homogeneous, central granites occur over
1409
The eastern lit-par-lit complex is composed of leucogranites, pegmatites, and aplites. It represents the earliest recognized intrusive phase of the batholith being intruded by the
homogeneous, megacrystic phases of the central Heerenveen
batholith in the northeast and bound and cross-cut by subvertical sheeted granites in the west (Fig. 3). The complex is best
preserved along the eastern margin of the batholith where the
granitoids intrude as foliation-parallel, cm- to m-wide sheets
parallel to the shallow to moderate SE dipping (25e45 )
gneissosity of the banded TTG country-rock gneisses (Table 1).
A similar complex is not developed along the western margin
of the batholith where the granitoids intrude subvertical basement gneisses (Fig. 4). The eastern lit-par-lit complex is a between 1 and 6 km wide zone and is exposed over a vertical
3.1. Eastern lit-par-lit complex
megacrystic granites. These granites form the volumetrically
dominant phases in the central parts of the batholith; (3) several 100 m to 2 km wide, ENE- to N-trending linear belts
made up of highly strained, subvertical sheeted granites that
bound the central megacrystic granites in the east and west.
These belts are compositionally the most heterogeneous zones
within the batholith and (4) late-to post-tectonic sheets and/or
plug-like, pink to grey homogeneous granites, mainly restricted to the SE parts of the batholith.
The spatial distribution, internal architecture and fabrics
within the four main phases are described below.
Fig. 2. Simplified geological map of the Heerenveen batholith intrusive into older basement granitoids and supracrustal greenstones. The northern and southern
contacts of the Heerenveen batholith are unconformably overlain by rocks of the Paleoproterozoic Transvaal Supergroup in the north and the Phanerozoic Karoo
Supergroup in the south. Form lines show the general trend and dips of the basement gneissosity and stretching lineations.
cross-cutting and fabric relationships, appear to have succeeded each other (Figs. 3 and 4). The first three phases
contain only locally developed magmatic, but pervasive
solid-state fabrics with regionally consistent ENE- to Ntrends, indicating the syntectonic timing of these granitoids.
The last granite phases cross-cut all earlier granitoids and
are devoid of solid-state fabrics. The classification of the fabrics present within the Heerenveen batholith into magmatic
and solid-state fabrics is based on the criteria outlined by
Paterson et al. (1989) and described in greater detail in
Belcher and Kisters (2006). The magmatic fabric in the Heerenveen batholith is defined by the preferred alignment of euhedral tabular K-feldspar megacrysts and is interpreted to signify
the re-orientation and alignment of phenocrysts normal to the
principal shortening direction during cooling and crystallization of the crystal mush, but with melt still present in the system (e.g. Paterson et al., 1998; Benn et al., 2001). The solidstate foliation is defined by the plastic deformation and elongation of minerals normal to the principal shortening direction
after cooling and crystallization below the solidus (e.g. Paterson et al., 1989).
The four main stages include (from old to young): (1)
sheet-like granitoids forming m-scale lit-par-lit injections
into the shallowly dipping basement gneisses, best preserved
along the eastern margin of the Heerenveen batholith. This
shallowly dipping sheeted complex, henceforth referred to as
the eastern lit-par-lit complex, can be traced for over 20 km
along strike (Fig. 3); (2) relatively homogeneous, commonly
1410
sheets dominate and constitute >80% of the outcrop. Significantly, isolated country-rock screens between the intrusive
granitoids retain their shallow SE dips with only little evidence
of rotation compared to country-rock gneisses outside the eastern lit-par-lit complex.
The shallowly dipping granitoid sheets contain a well developed, sheet-parallel, high-temperature, solid-state foliation,
particularly in the lower parts of the lit-par-lit complex, defined by the grain-shape preferred orientation of quartz and
quartzefeldspar aggregates and the orientation of phyllosilicates, mainly muscovite. Associated with the gneissosity is a
down-dip lineation defined by stretched quartz- and quartze
feldspar aggregates and muscovite. Both the solid-state foliation and stretching lineation in the intrusive sheets are parallel
to the planar and linear fabrics of the older country-rock
gneisses (Figs. 3c and 6a, b). It is clear that, on a regional
scale, the country-rock gneisses have acquired their fabric during the main phase of tectonism in the granitoid-greenstone
terrain at ca. 3230 Ga (Dziggel et al., 2002; Stevens et al.,
2002). This suggests an almost coaxial overprint of these older
Fig. 3. Geological map of the Heerenveen batholith showing: (a) The main intrusive phases and their distribution, reflecting the overall zonation of the batholith,
consisting of a central core of relatively homogeneous megacrystic granite, bound by compositionally heterogeneous marginal zones. (b) The four assembly stages
(stage 1 e oldest, to stage 4 e youngest) based on mainly cross-cutting relationships, but also fabric development and internal geometry. The correlation between
the compositionally different phases and their timing is detailed in Table 1. (c) The distribution of magmatic and solid-state fabrics in the batholith and gneissosities
in the surrounding basement. Note the magmatic foliation in the central, homogeneous megacrystic granite and the solid-state foliation and associated lineation in
the surrounding granites. The margin of the central megacrystic granites is bound by two synmagmatic shear zones and corresponds to the zones of heterogeneous
granite sheeting.
extent of ca. 350 m showing a crude vertical, internal zonation.
The base of the complex is characterized by isolated sheets
(1e2 cm and up to 2 m thick) that intrude parallel to the shallowly dipping gneissosity and compositional banding of the
TTG gneisses (Fig. 5a). This basal zone can be traced for
over 20 km at approximately the same elevation along the
eastern margin of the Heerenveen batholith. At this structural
level, intrusive sheets constitute between ca. 5 and 25% of the
outcrop. Most of the sheets are pegmatitic with up to 10 cm
large, euhedral K-feldspar crystals intergrown with quartz
and minor muscovite. Fine-grained aplites are also common,
while leucogranitic sheets are rare. At higher structural levels,
within ca. 100 m from the basal zone, granitic sheets become
more abundant and may constitute 50e60% of the outcrop.
The dm- to m-wide sheets form a branching and coalescing
network of foliation-parallel sills and cross-cutting low-angle
sheets that engulf rafts of the TTG basement (Fig. 5b). Fineto medium-grained leucogranites increase in abundance, while
pegmatites become subordinate. At the highest structural
levels exposed, within ca. 250 m from the basal zone, granitic
Pink granite II
Phases
Randomly orientated sheets and plug-like bodies
Occurrence
Undeformed (post-tectonic) cross-cutting pink granite I,
leucogranite I (centre) and TTG basement (southeast)
Age relationships
1411
Stage 4
Table 1
Summary of the main granite intrusive phases that compose the Heerenveen batholith
Stage 3
Granodioritic
dykes (not on Fig. 3)
Quartz monzonite
Predominantly intruded as a series of sheets
confined to within the synmagmatic shear zones
Megacrystic granite
Grey granite
Leucogranite II
Volumetrically the dominant phase forming a
large central homogeneous body
Limited outcroppings in south-central parts of
the batholith
Intruded as a series of sheets within and along
the margins of the synmagmatic shear zones
Pink granite I
Leucogranite I
Strong solid-state gneissosity. Intrudes into underlying TTG
basement. Intruded and crosscut by the megacrystic granite
Well-developed solid-state gneissosity (syntechtonic). Intruded
by the leucogranite, pink granite and quartz monzonite
(stage 3). Co-magmatic with megacrystic granite
Magmatic foliation superimposed by high-temperature
solid-state foliation (syntectonic). Intruded by the leucogranite
and pink granite (stage 3) along margins of the phase
High-temperature solid-state gneissosity (syntectonic).
Co-magmatic with pink granite I
High-temperature solid-state gneissosity (syntectonic).
Co-magmatic with pink granite I
Within the shear zones: high-temperature solid-state
gneissosity (syntectonic). Randomly orientated intrusions:
weak foliation to undeformed (late- to post-tectonic).
Both styles intrude into the megacrystic granite and
TTG basement
High-temperature solid-state gneissosity. Intrudes into
the megacrystic granite and TTG basement
Stage 2
Stage 1
Lit-par-lit intrusion of sheets and dykes found
predominantly along the eastern margin of the
batholith
cm- to m-wide leucogranite and pegmatite dykes that intrude
both the grey and megacrystic granitoids. The mainly subvertical dykes show predominantly N- (350e035 ) or ENE- (050e
090 ) trends. Subhorizontal leucogranite sheets also occur, but
are volumetrically subordinate (Belcher and Kisters, 2006).
The floor of the central megacrystic granitoids is not exposed.
However, basement gneisses exhibiting subhorizontal lithological and structural layering occurring along the eastern margin
of the granitoids may indicate the continuation of the shallowly
dipping basement gneisses from the east and below the central
parts of the Heerenveen batholith.
Closely packed feldspar megacrysts locally occur in cm- to
dm-wide bands that can be traced for several tens of metres,
outcrop permitting, defining a magmatic layering (Fig. 5c).
This layering is commonly steep and shows consistent
(E)NE trends across the central parts of the batholith (Figs.
3c and 6c). A magmatic foliation is regionally developed although it is relatively weak to absent in the central parts of
the megacrystic granitoids (Fig. 5d). The foliation is defined
by the preferred orientation of tabular feldspar megacrysts,
showing (E)NE trends in the southern and central parts of
the megacrystic granitoids, assuming more N-trends in the
north (Belcher and Kisters, 2006). The magmatic foliation
dips steeply to the SE and E. A high-temperature solid-state
foliation is subparallel to the magmatic foliation. The solidstate foliation is defined by elongate quartz-grain aggregates,
the grain-shape preferred orientation of quartzefeldspar
aggregates and the preferred orientation of phyllosilicates.
Cross-cutting pegmatite and aplite dykes are symmetrically
folded about the NE-trending fabric, indicating deformation
during mainly coaxial shortening (Belcher and Kisters,
Based on their relative age relationships and distinct emplacement styles the granites can be subdivided into four emplacement stages, as discussed in the text.
fabrics by 3.1 Ga fabrics recorded in the younger granitoid
sheets.
3.2. Central, megacrystic granites
The central parts of the Heerenveen batholith are made up
of relatively homogeneous K-feldspar megacryst-bearing
granitoids that underlie an area of ca. 200 km2 forming the
volumetrically dominant phases of the pluton (Fig. 3a). The
megacrystic granites are bounded and intruded by subvertical
granite sheets along their western and eastern margins, but are
intrusive into the eastern lit-par-lit complex (Fig. 4). Quartz,
K-feldspar, and plagioclase are the main components of the
mainly leucocratic granites with accessory amounts of muscovite, biotite, chlorite, apatite and zircon. The commonly euhedral and tabular-shaped K-feldspar megacrysts reach lengths
of up to 7 cm and show magmatic zoning. In places, quartz
forms rounded, up to 2 cm large, interstitial aggregates that result in a distinct studded weathering pattern of the rocks. The
K-feldspar megacrysts may either occur sporadically, as
evenly distributed phenocrysts, as irregularly shaped clusters,
or as trains of closely packed phenocryst aggregates. Transitions between megacryst-rich and megacryst-poor zones are
commonly gradational and intrusive relationships within the
megacrystic granitoids are only rarely observed. As such, contacts between different phases are cryptic or non-existent so
that the internal geometry and architecture of the central granites remains unclear. A fine- to medium-grained, but volumetrically subordinate homogeneous grey granite phase is
intrusive into the megacrystic granitoids, particularly in the
southern parts of the batholith. The only other intrusions are
1412
to >10 m and can be followed along strike for >100 m, outcrop permitting. Multiple sheet-in-sheet intrusions are common, indicated by low-angle intrusive contact relationships
(Fig. 7). Given the width of the subvertical sheeted domains
of up to 2 km, probably several hundred sheeting events are recorded within these domains. Northerly striking granite sheets
are confined to the northern and southern extents of the western margin of the Heerenveen batholith (Belcher and Kisters,
2006). Sheet-in-sheet intrusive contacts range from being
sharp to gradational, probably reflecting the relative time
gap between the full crystallization of earlier sheets and the
intrusion of subsequent magma batches. The domains of subvertical granite sheeting are compositionally the most heterogeneous zones within the Heerenveen batholith. Fine- to
medium-grained leucogranites are most common. Other phases
are, in decreasing order of abundance, pegmatites, fine-grained,
pink granites, quartz monzonites, megacrystic granites and
greyish granodiorites and diorites (Table 1, Fig. 3a). Along
the southern extent of the eastern sheeted domain, the steeply
dipping sheets envelop large rafts of banded, subhorizontal
basement gneisses. The basement gneisses are gently folded
Fig. 4. Simplified cross-sections of the western and eastern margins of the Heerenveen batholith highlighting the characteristic asymmetry of the intrusive phases of
the batholith shown in Fig. 3a.
2006). This solid-state foliation records a gradual increase in
strain intensity towards the western and, in particular, the eastern margin of the megacrystic granitoids. The higher fabric intensities are manifested by quartz ribbons and the formation of
a pervasive gneissosity that grades into protomylonitic textures. These highly strained textures along the margins of
the central megacrystic granitoids are associated with a distinct
change in intrusive style, and the homogeneous granitoids are
intruded and bounded by subvertical granite sheets.
3.3. Subvertical sheeted granites:
synmagmatic shear zones
Two ENE- to N-trending, curvilinear belts of subvertical
sheeted granites bound the eastern and western margins of
the megacrystic granitoids. These sheeted domains can be followed for over 25 km along strike ranging from ca. 500 m to
2e3 km in width (Fig. 3a).
The most characteristic feature of these zones is the abundance of subvertical, mainly ENE-trending granite sheets
(Fig. 6f). Individual sheets range in width from several cm
a
d
b
c
1413
with considerably higher strain intensities compared to the
wall-rocks or older sheets they intrude into (Fig. 8d). This is
a common feature also observed in other batholiths of the
GMS suite (Jackson and Robertson, 1983; Westraat et al.,
2005), indicating the strain localization into the intrusive
sheets. Since the high-strain textures are defined by solid-state
fabrics, strain localization must have continued during the
early stages of subsolidus cooling of the sheets. The features
described above underline the syntectonic emplacement of
the granite sheets and the positive feedback between deformation and melt-bearing zones (Brown and Solar, 1998; Vigneresse and Tikoff, 1999). This led Belcher and Kisters (2006) to
suggest that the linear belts of subvertical sheeted granites represent synmagmatic shear zones. Notably, the high-strain fabrics of the synmagmatic shear zones are not developed outside
the confines of the Heerenveen batholith (Fig. 3c).
Non-coaxial fabrics and shear-sense indicators are common
and include s-type clasts, and S-C and S-C0 fabrics as well as
bookshelf-structures in fractured feldspar megacrysts, particularly common in pegmatitic sheets. In the ENE-trending subvertical sheeted complexes, predominantly dextral strike-slip
kinematics are recorded. However, sinistral shear-sense indicators are also observed and may occur in close spatial association with dextral strike-slip indicators. The N-trending
segments along the western sheeted margin, in contrast, show
mainly sinistral strike-slip. As such, the ENE- and N-trending
belts form a conjugate set of synmagmatic shear zones (Belcher
and Kisters, 2006). Shear-sense indicators are only observed on
horizontal or near-horizontal sections, irrespective of the
plunge of the stretching lineations, which suggest that the
Fig. 5. Photographs of field relationships between intrusive phases of the Heerenveen batholith showing: (a) and (b) Lit-par-lit intrusion of leucogranite and pegmatite sheets along the shallow dipping gneissosity of the TTG basement in the east of the Heerenveen batholith forming the eastern lit-par-lit complex. (c) Plan
view of the NE-trending magmatic layering defined by variations in the number of K-feldspar megacrysts within the central megacrystic granites. (d) ENEtrending, steeply dipping magmatic foliation defined by the preferred orientation of tabular cm-long K-feldspar megacrysts from the central megacrystic granites.
into upright, NE- to N-trending, subhorizontal folds so that the
intrusive sheets have an axial planar orientation with respect to
the folds.
The second characteristic feature of these domains is the
development of pervasive solid-state fabrics, parallel to subparallel to the sheet margins (Fig. 6e, f). The fabric comprises
a high-temperature solid-state foliation and lineation. The
foliation is defined by quartz ribbons that alternate with recrystallised quartzefeldspar domains, locally resulting in the
formation of banded gneisses. Where present, the preferred
orientation of mica accentuates the foliation. Protomylonitic
and mylonitic fabrics are common and quartz ribbons may
show axial ratios of >20:1 in horizontal sections. Significantly, the sheet-parallel gneissosity of earlier sheets is truncated by subsequent sheets that intrude at low angles
(Fig. 7c, d) indicating the synmagmatic timing of fabric development. Large K-feldspar phenocrysts are marginally recrystallised and form augen-shaped mantled porphyroclasts
(Fig. 8a, b). The feldspar megacrysts also undergo brittle deformation shown by large bookshelf-type clasts that are transected by microfaults with quartz in the strain shadows.
Intrusive sheets may also be folded into m-scale intrafolial,
isoclinal folds, indicating progressive layer transposition
(Fig. 8c). Stretching lineations are defined by elongated
quartz- and quartzefeldspar aggregates. The plunge of the
stretching lineations ranges from subhorizontal to subvertical
even within individual sheets. The fabric intensity recorded
in individual sheets by e.g. the aspect ratios of quartz ribbons
or the degree of recrystallisation and grain-refinement is
highly variable. It is not uncommon to find intrusive sheets
1414
Two main features are pertinent for an understanding of the
assembly and emplacement controls of the Heerenveen batholith. Firstly, the Heerenveen batholith was assembled through
episodically emplaced magma batches, and at least four
main geometrically distinct intrusive stages can be identified.
Two of these show well-preserved sheeted geometries (stages
1 and 3). Cross-cutting relationships in these sheeted domains
record a multitude of sheeting events documenting the continued addition of melt to the batholith. Secondly, magmatic and
solid-state fabrics in the composite pluton point to its emplacement during regional NW-SE directed subhorizontal shortening (Belcher and Kisters, 2006). The penetrative fabrics
formed during crystallization of the different magma batches
and continued to develop during the early stages of subsolidus
cooling. On a regional scale, the fabrics in the batholith correspond to the NE trend of folds and thrusts in the Barberton
greenstone belt having formed during the D3 tectonism at
ca. 3.1 Ga (e.g. De Ronde and De Wit, 1994; Kamo and Davis,
1994). This underlines the regional extent of crustal shortening
and the syntectonic emplacement of the Heerenveen batholith.
The following discussion is presented with respect to the
above background.
4. Discussion
Post-kinematic granitoids truncate all earlier phases and are
devoid of any macroscopic fabrics. The mainly fine- to medium-grained, pink to greyish granitoids are confined to the
SE and E margin of the batholith (Fig. 3). In the SE, the granitoids abut sharply against the steeply dipping, NE-trending
supracrustal Schapenburg schist belt that is made up of amphibolites, ultramafic talc-carbonate schists, serpentinites and
minor metasediments. In the eastern exposures, the pink granitoids intrude the subvertical sheeted granite complex as seemingly randomly orientated sheets and plug-like bodies. They
sharply truncate the earlier fabrics and large, rotated rafts
within the pink granitoids show the typical sheeted nature
and penetrative gneissosities characteristic for the marginal
synmagmatic shear zones (Fig. 8f).
3.4. Post-kinematic granitoids
sheeted domain (Fig. 3a). The breccias are composed of
dm- to m-scale, angular fragments of megacrystic granites
intruded by subvertical leucogranite and granite sheets. The
intrusive sheets still show mainly ENE trends (parallel to those
in the synmagmatic shear zones) forming a network of branching and coalescing dykes (Fig. 8e). Locally, intrusive stockworks are developed. Up to six separate intrusive phases and
brecciation events can be identified on individual pavements,
testifying to the multiple intrusive relationships. Both the intrusive granites as well as the fragments of megacrystic granites contain the regionally developed ENE-trending solid-state
gneissosity, although at lower strain intensities compared to
the subvertical sheeted granites, also suggesting very little rotation of the fragments during brecciation.
Fig. 6. Lower hemisphere, equal area projections of the orientation of intrusive
and/or structural elements discussed in the text. (a) Great circles (n ¼ 23) to
the sills of the eastern lit-par-lit complex and poles to the gneissosity (crosses,
n ¼ 29) of the basement along the eastern margin of the Heerenveen batholith.
(b) Poles to high-temperature solid-state foliation (crosses, n ¼ 67) and mineral stretching lineation (dots, n ¼ 18) within leucogranite and aplite sheets
of the eastern lit-par-lit complex (stage 1) illustrating the shallow and sheetparallel dip of the foliation, parallel to the basement gneissosity. (c) Poles
(squares, n ¼ 17) to magmatic layering of the central megacrystic granites
(stage 2) illustrating the preferred NE trend. (d) Poles (triangles, n ¼ 43) to
the magmatic foliation of the central megacrystic granites (stage 2) mainly defined by the preferred orientation of euhedral K-feldspar megacrysts. (e) Poles
to the high-temperature solid-state foliation in megacrystic granites (stage 2)
and the bounding synmagmatic shear zone associated sheeted margins (stage 3)
(crosses, n ¼ 98). Dots (n ¼ 30) illustrate the considerable scatter of mineral
stretching lineations in the synmagmatic shear zones. (f) Great circles (n ¼
103) to intrusive sheets from the subvertical synmagmatic shear zones.
high-strain zones are transpressional shear zones. The conjugate orientation and shear sense are consistent with the subhorizontal, NW-SE directed shortening strain during the
emplacement of the Heerenveen batholith.
The contacts between the subvertical sheeted domains and
the central megacrystic granites are developed as intrusive
breccias. These breccia zones are up to 4 km wide, such as
along the contacts between the central granites and the eastern
a
c
b
d
1415
emplacement of the sills was aided by volatile-driven intrusion
and wall-rock translation (Weinberg and Searle, 1999). The
preservation of the original orientation of basement rafts and
screens within the injection complex corroborates the largely
non-rotational wall-rock translation during sheet emplacement. During ongoing shortening the initially subhorizontal
TTG gneisses and intrusive sills were folded about a NEtrending axis and steepened to moderate (25e45 ) SE dips,
i.e. into an orientation where maximum shear stresses would
be resolved during NW-SE directed, subhorizontal shortening
(Fig. 9b). This may explain the formation of the sheet-parallel
gneissosities and down-dip stretching lineations in the intrusive sheets that are confined to the sills of the eastern litpar-lit complex and that are not found in any other phases of
the Heerenveen batholith. In summary, both the sheet-like geometry and lit-par-lit style of emplacement argue for a strong
control of these earliest intrusive phases by pre-existing wallrock anisotropies and the orientation of the wall-rock structures with respect to regional strains.
Fig. 7. Photographs of field relationships between different intrusive phases of the Heerenveen batholith showing: (a) Whaleback outcrops, in the foreground, containing m-wide, steep easterly dipping sheets along the western margin of the Heerenveen batholith. The zone of sheeted granites along the western margin of the
Heerenveen batholith is up to 500 m wide, building up much of the seemingly massive granite slopes in the background. Field of view is ca. 30 m wide in the
foreground, looking towards north. (b) Heterogeneous zone of subvertical sheeted granites along the eastern margin of the central megacrystic granites. Sheets
are between 5 cm and 2 m wide and are continuous along strike across the outcrops for up to 200 m. Field of view ca. 15 m (foreground). (c) Centimetre-scale,
NE-trending granite and pegmatite sheets from the eastern subvertical sheeted margin (plan view). Low-angle cross-cutting relationships are common, testifying to
the multiple intrusion of the granitoid sheets; sheet margins are annotated for clarity. All sheets contain a pervasive solid-state foliation, subparallel to the sheet
margins. (d) Oblique plan view of the low-angle cross-cutting relationships between a foliated medium-grained leucogranite (centre of the photograph) and later,
medium-to coarse-grained, foliated granites bounding the central leucogranite; sheet margins annotated for clarity; A5 notebook for scale. This feature is common
throughout the sheeted margins and can be followed in outcrops for several hundred metres.
4.1. The progressive development of
emplacement controls
The emplacement of the sills of the early lit-par-lit complex
(stage 1) was determined by the gneissosity and lithological
banding of the TTG basement gneisses, suggesting that it
was mainly differences in the tensile strengths parallel to
and across the pre-existing anisotropies that controlled the localization of the sills (e.g. Brisbin, 1986; Lucas and St-Onge,
1995). In their present orientation, the sills and enveloping
TTG gneisses show shallow- to moderate-SE dips. It is conceivable that the sills were originally emplaced as subhorizontal sheets, parallel to the then subhorizontal basement
foliation, as is locally preserved in the more central parts of
the batholith. In this scenario, the sheets intruded along the
s1es2 plane during subhorizontal shortening, occupying tensile fractures in addition to being emplaced along the gneissosity (Fig. 9a). The abundance of pegmatites and, as such, the
volatile-rich nature of the intruding sheets suggest that the
1416
a
c
e
b
d
f
foliations is not unusual in granite plutons, and has been suggested to represent the transition from magmatic to solid-state
flow during the continued cooling and crystallization of the plutons during regional deformation. The presence of melt allows
for the rotation of the phenocrysts, without internal or marginal
solid-state deformation, while the interlocking crystals of the
Fig. 8. Photographs of field relationships between different intrusive phases of the Heerenveen batholith showing: (a) Intense marginal recrystallisation of large
feldspars separated by quartz ribbons from a pegmatite sheet illustrating the development of high-temperature, high-strain solid-state fabrics in intrusive sheets.
Quartz ribbons run at low angles through the sheet delineating C0 planes of a S-C0 fabric, indicating dextral sense of shear (plan view, eastern synmagmatic shear
zone). (b) K-feldspar phenocrysts fractured and displaced along a small-scale fault, occupied by a quartz stringer (qtz). The fault has the orientation of a C0 plane
indicating dextral sense of shear. Flattened quartz aggregates define a grain-shape preferred orientation (plan view). (c) Isoclinal folding of a pegmatite sheet (centre of photo) within subvertical granite sheets of the eastern bounding synmagmatic shear zone (oblique plan view, A5 notebook for scale). (d) Subvertical sheet-insheet intrusion (plan view) from the eastern synmagmatic shear zone. This illustrates the strong, sheet margin-parallel gneissosity defined by positively weathering
quartz ribbons (central granite sheet) intruded into a finer-grained leucogranite sheets (bottom and top of photo) and the effects of strain partitioning into the intrusive sheet. (e) Metre-scale intrusive breccia of medium-grained leucogranite intruding coarse-grained megacrystic granite. These breccias are developed over
widths of up to 4 km. Both the megacrystic granites (stage 2) and the intruded leucogranite sheets (stage 3) contain a well developed NE-trending gneissosity,
running approximately parallel to the top to the photo (oblique plan view, eastern breccia zone, annotations for clarity). (f) Metre-scale rotated rafts of sheeted
leucogranites (annotated by dashed lines) and granites (stage 3) intruded by late-stage undeformed pink granites (stage 4) in the northeastern margin of the Heerenveen batholith.
Only little can be deduced about the emplacement controls or
even the geometry of the central megacrystic granites, the next
younger phases of the Heerenveen batholith (stage 2). Intrusive
contacts within the granites and wall-rock relationships are only
rarely exposed and the granites are commonly massive in appearance. The parallelism between magmatic and solid-state
Fig. 9. Schematic diagrams showing the incremental construction of the Heerenveen batholith through the four main intrusive stages outlined in the text: (a) Stage 1:
cross-sectional view showing the intrusion of the early leucogranite, aplite and pegmatite sheets along the subhorizontal gneissosity of the TTG basement during
subhorizontal NW-SE shortening. The emplacement is controlled by the regional strain and pre-existing wall-rock anisotropies of the basement. (b) Folding of the
basement about a NW-SW orientated axis leads to the rotation of the eastern lit-par-lit complex. During this stage, the sills develop the layer-parallel solid-state
foliation and down-dip lineation, while low-angle cross-cutting dykes testify to the continued granite sheeting. (c) Stage 2: The formation of the homogeneous
central granites as steady-state magma chambers is envisaged to have been facilitated by the thermal ground preparation provided by the earlier lit-par-lit intrusions. (d) Map view showing the central megacrystic granites truncating the wide eastern margin of the eastern lit-par-lit complex. (e) Stage 3: Following the
formation of a steady-state magma chamber (stage 2), strain localization and partitioning along the margins of the central magma chamber during the regional
NW-SE subhorizontal contraction leads to the initiation and development of the bounding synmagmatic shear zones. This results in the profound change in
the construction style of the batholith, from predominantly wall-rock anisotropy controlled emplacement (stage 1) to melt controlled emplacement (stage 3).
The shear zones represent melt conduits for melt ascent which fed and inflated the now eroded, higher structural levels of the batholith. (f) Map view of stage
3 showing the temporal and spatial link between the shear zones and the heterogeneous granite sheeting along the margins of the megacrystic granite. (g) Stage
4: The final, post-tectonic stage in the construction of the batholith sees the emplacement of randomly orientated, smaller sheet- and plug-like intrusions. The sharp
termination of the late-stage intrusive against wall rocks, suggests an emplacement controlled by pre-existing anisotropies. (h) Map view of stage 4.
1418
least parts thereof, were partially molten for the bounding
shear zones to nucleate.
Granite-sheeting in the km-wide synmagmatic shear zones
and adjoining breccia zones have contributed to the progressive construction of the Heerenveen batholith. However, the
abundance and subvertical orientation of the shear zone-parallel sheets suggest that the synmagmatic shear zones, at their
present level of exposure, have most likely represented melt
transfer zones and ascent conduits, rather than the final emplacement sites of the granites. Melt transfer and the positive
feedback between melt transfer zones and deformation, i.e.
shear zones, is a widely documented feature (e.g. Hutton,
1982; McCaffrey, 1992; Brown and Solar, 1998). Recent
models for magma transfer in transpressional and contractional shear zones invoke magma overpressuring as the main
driving force for melt ascent. Magma overpressuring occurs
as a result of the buoyancy of the melt, the increase of vapour
pressures during the late stages of melt ascent and tectonic
pressures, i.e. deviatoric stresses in actively deforming environments (e.g. Ingram and Hutton, 1994; Hogan and Guilbert,
1995; De Saint-Blanquat et al., 1998). Significantly, the upward movement of melt is facilitated by vertical pressure gradients that are greatest in vertical structures, i.e. structures
representing the shortest connection to the free boundary of
Earth’s surface. Thus, the mainly shallowly dipping basement
gneissosity that determined the emplacement and orientation
of earlier phases of the Heerenveen batholith had an unfavourable orientation for a buoyancy-driven melt ascent, rather
leading to the ponding of magmas along the pre-existing anisotropies. The subvertical synmagmatic shear zones, in contrast, provided favourably inclined conduits for a vertical
melt transfer. The subvertical orientation of the shear zones
also means that a potentially larger variety of sources may
have been tapped in the subhorizontal basement complex,
which is possibly reflected in the fact that the geochemical heterogeneity previously documented for the Heerenveen batholith (e.g. Anhaeusser et al., 1983; Yearron, 2003) is almost
exclusively confined to the subvertical sheeted domains.
crystallizing magma form a load-bearing framework that is sufficiently strong to transmit regional stresses (Vigneresse and
Tikoff, 1999). Progressive deformation after full crystallization
of the granite then leads to the formation of high-temperature
solid-state fabrics (e.g. Paterson et al., 1989). The subvertical
orientation of both the magmatic and solid-state foliations
throughout the Heerenveen batholith tracking the shortening
(XY-) plane of the finite strain ellipsoid, is consistent with the
emplacement and cooling of the pluton during ongoing subhorizontal shortening (Fig. 9c, d).
The next main stage in the assembly of the Heerenveen
batholith (stage 3) is represented by the subvertical sheeted domains along the margins of the central granites, and is associated with striking changes in the emplacement controls of
subsequent granite phases (Fig. 9e, f). The discrete, multiple
granite sheets contained in the synmagmatic shear zones no
longer exhibit any controls by wall-rock anisotropies that
have largely determined the emplacement of earlier phases.
In contrast, the multiple additions of melts to the marginal
zones of batholith points to the significance of earlier magmas
for the localization of subsequent melt batches. In other words,
at this advanced stage in the assembly of the Heerenveen batholith it is factors intrinsic to the pluton that govern the emplacement of subsequent granite phases, while the influence
of external controls, primarily wall-rock anisotropies, becomes
subordinate (Fig. 10).
Both experimental and field studies (e.g. Grujic and Mancktelow, 1998; Walte et al., 2005) have documented the nucleation and progressive growth of conjugate shear zones around
rheologically weaker melt-bearing zones during coaxial shortening of mixed rock-melt systems. The conjugate sets of noncoaxial shear zones typically surround lower strained pods that
record near-coaxial pure shear deformation. This deformation
pattern closely mimics the location, conjugate arrangement
and kinematics of the bounding synmagmatic shear zones
around the central megacrystic granites of the Heerenveen
batholith (Belcher and Kisters, 2006). One of the prerequisites
for this to occur is that the central megacrystic granites or at
Fig. 10. Diagrammatic synopsis of the changing emplacement styles and controls during the incremental construction of the Heerenveen batholith. Controlling
factors can be grouped into pre-existing factors unrelated to the pluton itself (external) and those related to the pluton, i.e. melt related (internal). Early intrusive
phases are predominantly controlled by the wall-rock anisotropy and the regional strain field (external factors). With the introduction of more melt, the influence of
the wall-rock anisotropy is reduced, and conversely, controls by previous intruded melts and the effects of strain localization determine the emplacement of subsequent melt batches (internal factors). In the absence of a deviatoric stress field, post-tectonic granite factors are again controlled by the wall-rock anisotropy.
1419
sheeted architecture, particularly at lower structural levels of
the lit-par-lit complex pointing to the relatively quick cooling
of the narrow intrusions below their solidus. Given a sufficiently high magma supply rate and a high-rate of sheeting,
the repeated addition of heat associated with multiple intrusions will increase the ambient wall-rock temperature
(Furlong et al., 1991). Continued sheeting, preserved by the
coalescing sills and low-angle dykes at higher structural levels
in the eastern lit-par-lit complex will result in slower cooling
rates of newly added sheets as the ambient wall-rock temperatures will increase. This has the effect that larger steady-state
magma chambers may be constructed through the continued
addition of relatively small melt batches, provided that the ambient temperatures are elevated above the intrusions’ solidi
(e.g. Hanson and Glazner, 1995; Fleck et al., 1996). The formation of a steady-state magma chamber by this process of
progressive sheeting also implies that the crystallization front
within the chamber may migrate (Marsh, 1996; Yoshinobu
et al., 1998), depending on the site of new magma additions.
As a consequence, the crystallization front will not necessarily
track the original geometry of the granites, so that the intrusive
contacts between different granite phases may be homogenized and obliterated. This leads to the commonly observed
cryptic contacts in the centres of large granite batholiths
(e.g. Glazner et al., 2004). A similar process is invoked here
for the development of the central megacrystic granites, which
lack clear internal contacts. Thus, the early stage of multiple
granite sheets preserved in the several km-wide lit-par-lit complex along the eastern margin of the Heerenveen batholith represent the thermal ground preparation for the development of
the subsequent central, more massive megacrystic phases.
The introduction of the rheologically weaker steady-state
magma chambers, in turn, is a prerequisite for the nucleation
of the bounding synmagmatic shear zones, which are similarly
documented for the adjoining Mpuluzi batholith (Jackson and
Robertson, 1983; Westraat et al., 2005). The synmagmatic
shear zones, in turn, not only contribute to the incremental
construction of the batholith, but also to the thermal insulation
and maintenance of a central steady-state magma chamber, by
providing a buffer between the central granites and the surrounding cold wall rocks. The degree of preservation of the internal sheeted geometry in the synmagmatic shear zones
indicates that crystallization rates of individual sheets were
faster than the emplacement rates (compared to the central
megacrystic granites where crystallization rates were slower
than emplacement rates). This inhibited post-emplacement
textural or chemical homogenization along the boundaries between the cooler wall rocks and the central granites, preserving the distinct sheeted nature of granitic intrusive phases.
The example of the 3.1 Ga Heerenveen batholith illustrates
how emplacement controls of syntectonic plutons undergo
progressive adjustments in response to the incremental construction and successive addition of melt batches. The emplacement controls can be conceptualized into external
factors, i.e. those that are independent of the magmas, and internal factors intrinsic to the magmas (Fig. 10). The latter involve mainly rheological changes as a consequence of granite
Alternative interpretations of the synmagmatic shear zones
as, e.g. bounding structures that accommodate the volume of
the intruding magma and inflation of the pluton (Tobisch
and Cruden, 1995; Cruden, 1998) or as already present regional shear zones that guided and facilitated melt migration
or emplacement (e.g. Brown and Solar, 1998) appear unlikely.
The sinistral and dextral transpressive kinematics of the conjugate shear zones record the regional shortening, rather than accommodating wall-rock displacement as a result of either roof
uplift, floor depression or the lateral translation of wall rocks
(e.g. Cruden, 1998; Cruden and McCaffrey, 2001). Similarly,
regionally developed shear zones with comparable orientations and kinematics are not found in the surrounding basement gneisses and the high-strain fabrics are only developed
within the confines of the Heerenveen batholith.
The lack of magmatic and solid-state fabrics in the final intrusive phases of the Heerenveen batholith is interpreted to indicate their post-tectonic timing (stage 4). These granites
intrude the earlier phases of the Heerenveen batholith as seemingly, randomly orientated sheets or plugs, in stark contrast to
the dominant ENE- and N-trends of granites in large parts of
the pluton (Fig. 9g, h). However, in places where the late-stage
granites intrude and are in contact with the surrounding basement rocks, pre-existing wall-rock anisotropies exert a strong
control on the geometry of the granites. This is the case in the
SE parts of the Heerenveen batholith, where the fine-to medium-granites sharply terminate against the metasediments
and metavolcanics of the schist belt. The post-tectonic phases
also indicate the prolonged intrusion history and that magmatic activity outlasted the regional D3 tectonism.
4.2. The role of sheeting for pluton construction and
steady-state magma chamber formation
The pattern of marginal zones of sheeted granites enveloping cores of more massive granites is typical for most of the
GMS suite plutons in the Barberton granitoid-greenstone terrain (Anhaeusser and Robb, 1983). On regional maps, this zonation is delineated by what was then referred to as the
‘‘marginal migmatite zones’’ surrounding the main plutons
of e.g. the Heerenveen, Mpuluzi or Nelspruit batholith (e.g.
Anhaeusser et al., 1981; Anhaeusser and Robb, 1983; Robb
et al., 1983). We suggest that this zonal distribution of granite
phases with distinct internal architectures and geometries and
their relative timing with respect to each other outlined in this
study, point to some of the underlying mechanisms through
which the large batholiths of the GMS suite were emplaced.
A number of recent works that have modelled the thermal evolution and emplacement of large plutonic complexes at shallow-crustal levels emphasize the significance of sheeted
margins for the construction of steady-state magma chambers
(Furlong and Myers, 1985; Hanson and Glazner, 1995; Yoshinobu et al., 1998). Considering the shallow emplacement level
of the Heerenveen batholith (Hunter, 1973; Anhaeusser and
Robb, 1983; Robb et al., 1983), the earliest intrusions of the
eastern lit-par-lit complex were intruded into presumably
cool wall rocks. This corresponds to the well-preserved
1420
Anhaeusser, C.R., Robb, L.J., 1983. Geological and geochemical characteristics
of the Heerenveen and Mpuluzi batholiths south of the Barberton greenstone
belt and preliminary thoughts on their peterogenesis. In: Anhaeusser, C.R.
(Ed.), Contributions to the Geology of the Barberton Mountain Land. Special Publication of the Geological Society of South Africa 9, 131e152.
Anhaeusser, C.R., Robb, L.J., Viljoen, M.J., 1981. Provisional geological map
of the Barberton greenstone belt and surrounding granitic terrane, eastern
Transvaal and Swaziland: Geological Society of South Africa, Johannesburg. Scale 1:250000.
Anhaeusser, C.R., Robb, L.J., Barton Jr., J.M., 1983. Mineralogy, petrology
and origin of the Boesmanskop Syeno-granite complex, Barberton Mountain Land, South Africa. In: Anhaeusser, C.R. (Ed.), Contributions to the
Geology of the Barberton Mountain Land. Special Publication of the Geological Society of South Africa 9, 169e184.
Archanjo, C.J., Fetter, A.H., 2004. Emplacement setting of the granite sheeted
pluton of Esperança (Brasiliano orogen, Northeastern Brazil). Precambrian
Research 135, 193e215.
Belcher, R.W., Kisters, A.F.M. Emplacement of the Heerenveen batholith
along synmagmatic shear zones: evidence for regional-scale shortening
during craton-scale transtensional tectonics, Barberton granite-greenstone
terrain, South Africa. Geological Society of America. Special paper 405,
211e232.
Benn, K., Paterson, S.R., Lund, S.P., Pignotta, G.S., Kruse, S., 2001. Magmatic
fabrics in batholiths as markers of regional strains and plate kinematics:
example of the Cretaceous Mt. Stuart batholith. Physics and Chemistry
of the Earth, Part A: Solid Earth and Geodesy 26, 343e354.
Bergantz, G.W., 1991. Physical and chemical characterization of plutons. In:
Kerrick, D.M. (Ed.), Contact Metamorphism. Reviews in Mineralogy,
vol. 26. Mineralogical Society of America, pp. 13e42.
Brisbin, W.C., 1986. Mechanics of pegmatite intrusion. American Mineralogist 71, 644e651.
Brown, M., Solar, G.S., 1998. Granite ascent and emplacement during contractional deformation in convergent orogens. Journal of Structural Geology
20, 1365e1393.
Coleman, D.S., Gray, W., Glazner, A.F., 2004. Rethinking the emplacement and
evolution of zoned plutons: geochronological evidence for incremental assembly of the Tuolomne Intrusive Suite, California. Geology 32, 433e436.
Cruden, A.R., 1990. Flow and fabric development during the diapiric rise of
magma. Journal of Geology 98, 681e698.
Cruden, A.R., 1998. On the emplacement of tabular granites. Journal of the
Geological Society 155, 853e862.
Cruden, A.R., McCaffrey, K.J.W., 2001. Growth of plutons by floor subsidence:
implications for rates of emplacement, intrusion spacing and meltextraction mechanisms. Physics and Chemistry of the Earth, Part A: Solid
Earth and Geodesy 26, 303e315.
D’Lemos, R.S., Brown, M., Strachan, R.A., 1992. Granite magma generation,
ascent and emplacement within a transpressional orogen. Journal of the
Geological Society, London 149, 487e490.
Davidson, C., Hollister, L.S., Schmid, S.M., 1994. Role of melt in the formation of a deep-crustal compressive shear zone: the MacLaren Glassier
metamorphic belt, south-central Alaska. Tectonics 11, 348e359.
De Ronde, C.E.J., Kamo, S., Davis, D.W., De Wit, M.J., Spooner, E.T.C.,
1991. Field, geochemical and UePb isotopic constraints from hypabyssal
felsic intrusions within the Barberton greenstone belt, South Africa: implications for tectonics and the timing of gold mineralization. Precambrian
De Ronde, C.E.J., De Wit, M.J., 1994. Tectonic history of the Barberton greenstone belt, South Africa: 490 million years of Archean crustal evolution.
Tectonics 13, 983e1005.
De Saint-Blanquat, M., Tikoff, B., Teyssier, C., Vigneresse, J.L., 1998. Transpressional kinematics and magmatic arcs. In: Holdsworth, R.E.,
Strachan, R.A., Dewey, J.F. (Eds.), Continental Transpressional Tectonics.
Geological Society of London, Special Publication 135, 327e340.
Dziggel, A., Stevens, G., Pojoul, M., Anhaeusser, C.R., Armstrong, R.A.,
2002. Metamorphism of the granite-greenstone terrane south of the Barberton greenstone belt, South Africa: an insight in the tectono-thermal evolution of the ‘lower’ portions of the Onverwacht Group. Precambrian
plutonism, either as a result of the introduction of rheological
heterogeneities in the form of melts, associated strain localization and strain partitioning as well as wall-rock heating during
granite intrusion.
5. Conclusions
The emplacement of the successive phases of the Heerenveen batholith records a positive feedback loop between
melt emplacement and deformation, resulting in transient
changes in the controls and styles of emplacement of the
magmas. The earliest phases of pluton emplacement are
largely controlled by external factors, mainly determined by
the regional strain field and the presence and orientation of
pre-existing anisotropies. During the progressive assembly of
the pluton, the emplacement of subsequent phases is increasingly influenced by factors intrinsic of the magmas
(Fig. 10). Wall rocks are heated by continued granite sheeting
to the stage when steady-state magma chambers can be established. At this stage, strain localization and partitioning around
the steady-state magma chambers become the predominant
factor. Melt transfer and emplacement are controlled by synmagmatic shear zones that nucleated around the central
magma chambers. The thermal insulation of the steady-state
magma chambers through continued addition of melt and
heat to the bounding shear zones may lead to the homogenization of the originally heterogeneous central zones. This contributes to the development of the commonly observed
homogeneous, massive granites in the central parts of the batholith, as is predicted through thermal modelling studies of incrementally constructed magma chambers (Yoshinobu et al.,
1998). The fact that numerous large batholiths of the GMS
suite show similar zonal architecture of marginal sheeted
phases and more massive cores, suggests that similar processes
to those described here may have a wider application to the
construction of sheeted batholiths.
Acknowledgements
The material is based upon work supported by a South
African National Research Foundation (NRF) grant awarded
to Alex Kisters (GUN no. 2053186). Richard Belcher acknowledges financial support via the NRF and the University
of Stellenbosch towards a Post-doctoral Fellowship. The authors greatly appreciate the access to lands and the hospitality
of farmers and residents in and around the town of Badplaas
during fieldwork. We thank J.-F. Moyen for comments on an
earlier version of the manuscript and K. Benn and J.-L.
Bouchez for their helpful reviews.
References
Anhaeusser, C.R., 1983. The geology of the Schapenburg Greenstone Remnant
and surrounding Archaean granitic terrane south of Badplaas, Eastern
Transvaal. In: Anhaeusser, C.R. (Ed.), Contributions to the Geology of
the Barberton Mountain Land. Special Publication of the Geological Society of South Africa 9, 31e45.
1421
Marsh, B.D., 1996. Solidification fronts and magmatic evolution. Mineralogy
Magazine 60, 5e40.
McCaffrey, K.J.W., 1992. Igneous emplacement in a transpressive shear zone.
Ox Mountains igneous complex. Journal of the Geological Society of London 149, 221e235.
Miller, R.B., Paterson, S.R., 2001. Construction of mid-crustal sheeted plutons:
examples from the North Cascades, Washington. Geological Society of
America Bulletin 113, 1423e1442.
Neves, S.P., Vauchez, A., Feraud, G., 2000. Tectono-thermal evolution, magma
emplacement, and shear zone development in the Caruara area (Borborema
Province, NE Brazil). Precambrian Research 99, 1e32.
Paterson, S.R., Vernon, R.H., Tobisch, O.T., 1989. A review of criteria for the
identification of magmatic and tectonic foliations in granitoids. Journal of
Structural Geology 11, 349e363.
Paterson, S.R., Fowler Jr., T.K., Schmidt, K.L., Yoshinobu, A.S., Yuan, E.S.,
Miller, R.B., 1998. Interpreting magmatic fabric patterns in plutons. Lithos
44, 53e82.
Robb, L.J., Anhaeusser, C.R., Van Nierop, D.A., 1983. The recognition of the
Nelspruit batholith north of the Barberton greenstone belt and its significance in terms of Archaean crustal evolution. In: Anhaeusser, C.R.
(Ed.), Contributions to the Geology of the Barberton Mountain Land. Special Publication of the Geological Society of South Africa 9, 117e130.
Rutter, E.H., Neumann, D.H.K., 1995. Experimental deformation of partially
molten Westerly granite under fluid-absent conditions, with implications
for the extraction of granitic magma. Journal of Geophysical Research
100, 15697e15715.
Stevens, G., Droop, G.T.R., Armstrong, R.A., Anhaeusser, C.R., 2002. Amphibolite facies metamorphism in the Schapenburg schist belt: a record
of mid-crustal response to w3.23 Ga terrane accretion in the Barberton
greenstone belt. South African Journal of Geology 105, 271e284.
Tikoff, B., Teyssier, C., 1994. Strain modeling of displacement field partitioning in transpressional orogen. Journal of Structural Geology 16, 1575e
1588.
Tobisch, O.T., Cruden, A.R., 1995. Fracture-controlled magma conduits in an
obliquely convergent continental magmatic arc. Geology 23, 941e944.
Tommasi, A., Vauchez, A., Daudré, B., 1995. Initiation and propagation
of shear zones in a heterogeneous continental lithosphere. Journal of
Geophysical Research 100, 22083e22101.
Vigneresse, J.L., Tikoff, B., 1999. Strain partitioning during melting and crystallizing felsic magmas. Tectonophysics 312, 117e132.
Walte, N.P., Bonsand, P.D., Passchier, C.W., 2005. Deformation of meltbearing systems e insight from in situ grain-scale analogue experiments.
Journal of Structural Geology 27, 1666e1679.
Weinberg, R.F., Searle, M.P., 1999. Volatile-assisted intrusion and autometasomatism of leucogranites in the Khumbu Himalaya, Nepal. Journal of
Geology 107, 27e48.
Westraat, J.D., Kisters, A.F.M., Poujol, M., Stevens, G., 2005. Transcurrent
shearing, granite sheeting and the incremental construction of the tabular
3.1 Ga Mpuluzi batholith, Barberton granite-greenstone terrain, South
Africa. Journal of the Geological Society of London 162, 373e388.
Wiebe, R.A., Collins, W.J., 1998. Depositional features and stratigraphic sections in granitic plutons: implications for the emplacement and crystallization of granitic magma. Journal of Structural Geology 20, 1273e1289.
Yearron, L.M., 2003. Archean granite petrogenesis and the implications for the
evolution of the Barberton Mountain Land, South Africa. Unpublished
Ph.D. Thesis, Kingston University, United Kingdom, 315 pp.
Yoshinobu, A.S., Okaya, D.A., Paterson, S.R., 1998. Modeling the thermal
evolution of fault-controlled magma emplacement models: implications
for the solidification of granitoid plutons. Journal of Structural Geology
20, 1205e1218.
Fitch, J.T., 1972. Plate convergence, transcurrent faults, and internal deformation adjacent to Southeast Asia and the western Pacific. Journal of
Geophysical Research 77, 4432e4460.
Fleck, R.J., Kistler, R.W., Wooden, J.L., 1996. Geochronological complexities
related to multiple emplacement history of the Tuolumne intrusive suite,
Yosemite National Park, California. Geological Society of America, Abstracts with Programs 28, 65.
Furlong, K.P., Hanson, R.B., Bowers, J.R., 1991. Modeling thermal regimes.
In: Derrick, D.M. (Ed.), Contact Metamorphism. Review in Mineralogy,
vol. 26, pp. 437e505.
Furlong, K.P., Myers, J.D., 1985. Thermalemechanical modeling of the role of
thermal stresses and stoping in magma contamination. Journal of Volcanology and Geothermal Research 24, 179e191.
Gerbi, C., Johnson, S.E., Paterson, S.R., 2004. Implications of rapid, dike-fed
pluton growth for host-rock strain rates and emplacement mechanism.
Journal of Geology 26, 583e594.
Glazner, A.F., Bartley, J.M., Coleman, D.S., Gray, W., Taylor, R.Z., 2004. Are
plutons assembled over millions of years by amalgamation from small
magma chambers? GSA Today 14, 4e11.
Grujic, D., Mancktelow, N.S., 1998. Melt-bearing shear zones: analogue experiments and comparison with examples from southern Madagascar. Journal of Structural Geology 20, 673e680.
Hanson, R.B., Glazner, A.F., 1995. Thermal requirements for extensional emplacement of granitoids. Geology 23, 213e216.
Hogan, J.P., Guilbert, M.C., 1995. The A-type Mount Scott granite sheet: importance of crustal magma traps. Journal of Geophysical Research 100,
15779e15792.
Hunter, D.R., 1973. The granitic rocks of the Precambrian in Swaziland. Special Publication of the Geological Society of South Africa 3, 131e145.
Hutton, D.H.W., 1982. A tectonic model for the emplacement of the main
Donegal granite. Journal of the Geological Society of London 139,
615e631.
Ingram, G.M., Hutton, D.H.W., 1994. The Great Tonalite Sill: emplacement
into a contractional shear zone and implications for Late Cretaceous to
early Eocene tectonics in southeastern Alaska. Bulletin of the Geological
Society of America 106, 715e720.
Jackson, M.P.A., Robertson, D.I., 1983. Regional implications of earlyPrecambrian strains in the Onverwacht Group adjacent to the Lochiel
Granite, north-west Swaziland. In: Anhaeusser, C.R. (Ed.), Contributions
to the Geology of the Barberton Mountain Land. Special Publication of
the Geological Society of South Africa 9, 45e62.
Johnson, S.E., Fletcher, J.M., Fanning, C.M., Vernon, R.H., Paterson, S.R.,
Tate, M.C., 2003. Structure, emplacement and in-situ expansion of the
San José tonalite pluton, Peninsular Ranges batholith, Baja California,
México. Journal of Structural Geology 25, 1933e1957.
Kamo, S.L., Davis, D.W., 1994. Reassessment of Archean crustal development
in the Barberton Mountain Land, South Africa, based on UePb dating.
Tectonics 13, 167e192.
Karlstrom, K.E., Miller, C.F., Kingsbury, J.A., Wooden, J.L., 1993. Pluton emplacement along an active ductile thrust zone, Piute Mountains, southeastern California: interaction between deformational and solidification
processes. Geological Society of America Bulletin 104, 176e191.
Lucas, S.B., St-Onge, M.R., 1995. Syn-tectonic magmatism and the development of compositional layering, Ungava Orogen (northern Quebec, Canada). Journal of Structural Geology 17, 475e491.
Mahan, K.H., Bartley, J.M., Coleman, D.S., Glazner, A.F., Carl, B.S., 2003.
Sheeted intrusion of the synkinematic McDoogle pluton, Sierra Nevada,
California. Geological Society of America Bulletin 115, 1570e1582.
Marsh, B.D., 1982. On the mechanics of igneous diapirism, stoping and zone
melting. American Journal of Science 282, 808e855.
Journal of the Geological Society, London, Vol. 162, 2005, pp. 373–388. Printed in Great Britain.
Transcurrent shearing, granite sheeting and the incremental construction of the
tabular 3.1 Ga Mpuluzi batholith, Barberton granite–greenstone terrane,
South Africa
JA N U S D. W E S T R A AT 1, A L E X A N D E R F. M . K I S T E R S 1 , M A R C P O U J O L 2 & G A RY S T E V E N S 1
1
(e-mail: [email protected])
Department of Earth Sciences, Memorial University of Newfoundland, St. John’s, Nfld., A1B 3X5, Canada
2
Abstract: Structural, petrographic and geochronological studies show that the tabular 3.1 Ga Mpuluzi
batholith in the Barberton granite–gneiss terrane in South Africa was emplaced via a combination of external
and internal processes. External structural controls are indicated by systematic variations in intrusive
relationships and strain along the margins of the Mpuluzi batholith and are consistent with an emplacement of
the granite in a dilational jog within a NE–ENE-trending system of dextral transcurrent synmagmatic shear
zones. Internally, the Mpuluzi batholith is essentially made up of granite sheets. The structurally higher parts
of the granite are made up of shallowly dipping sheets that are underlain by an anastomosing network of
steeply dipping, variably deformed dykes and sheets. These granite sheets at lower structural levels intruded
either into the actively deforming shear zones or into extensional sectors between and along the bounding
shear zones. Multiple intrusive relationships and geochronological evidence suggests that granite sheeting and
the assembly of the pluton occurred over a period of 3–13 Ma. The spatial and temporal relationship between
deformation and magma emplacement reflects episodes of incremental dilation related to deformation along
the bounding shear zones and granite sheeting. The transition to the mainly subhorizontal granite sheets at
higher structural levels of the tabular Mpuluzi batholith indicates the intrusion of the granites during
subhorizontal regional shortening, where the reorientation of the minimum normal stress to vertical attitudes
at the shallow levels of emplacement allowed for vertical dilation and subhorizontal emplacement of the
granite sheets.
crustal differentiation was the Archaean, when significant parts
of the present continents were formed during accretionary
tectonic events and associated short-lived but voluminous episodes of granitoid magmatism. The actual nature of events that
prompted the production of these vast amounts of granitoids is as
ambiguous and controversial as the modes of emplacement of the
granitoid magmas. It is, thus, not surprising that Archaean
cratons and their granite–greenstone terranes have often been at
the centre of the debate about granite ascent and emplacement
mechanisms (e.g. Ramsay 1989; Jelsma et al. 1993; Ridley et al.
1997; Van Kranendonk et al. 2004). The Palaeo- to Mesoarchaean Barberton granite–greenstone terrane in the Kaapvaal
Craton in South Africa (Fig. 1) has featured very prominently in
this debate (Viljoen & Viljoen 1969; Anhaeusser 1973; De Wit
et al. 1992). This composite granite–greenstone terrane was
assembled during several tectonomagmatic episodes between c.
3.5 and 3.1 Ga (e.g. Anhaeusser & Robb 1980; Robb &
Anhaeusser 1983; Armstrong et al. 1990; De Ronde & De Wit
1994; Kamo & Davis 1994). Earlier, c. 3.5–3.2 Ga plutonic
suites are characterized by trondhjemites, tonalites and granodiorites. These rocks, collectively referred to as the TTG suite,
form typically relatively small (,100 to c. 500 km2 ) and almost
invariably gneissose bodies with largely concordant contact
relationships with the supracrustal greenstones. These features
have been explained by: (1) the diapiric ascent and emplacement
of the TTGs (e.g. Viljoen & Viljoen 1969; Anhaeusser 2001); (2)
the synkinematic, shallow-crustal underplating of the TTG suite
at the base of the largely allochthonous and thrust greenstone
Keywords: Archaean, Mpuluzi batholith, granites, shear zones, absolute age.
The petrogenesis of granites is almost invariably linked to active
orogenic settings and the transport and emplacement of granitic
magmas is now widely recognized to be aided and/or controlled
by regional-scale structures such as fault and shear zones, fold
structures or regional fabric patterns (e.g. Hutton 1988; Paterson
& Fowler 1993; Collins & Sawyer 1996; Clemens et al. 1997;
Petford et al. 2000). The intrusion of granitoids along and into
actively deforming wall rocks presents an elegant solution to the
so-called space problem of granite emplacement in that deformation potentially creates regions of localized dilation in a variety
of kinematic scenarios, including extensional, convergent and
wrench-tectonic environments (e.g. Guineberteau et al. 1987;
Hutton & Ingram 1992; Tikoff & Teyssier 1992; Vauchez et al.
1997; Brown & Solar 1998). One of the most widely used
approaches to decipher the actual mechanisms of granite emplacement is the structural analysis of wall-rock strains in the strain
aureole of granites and within the granites themselves (Paterson
et al. 1989; Ramsay 1989). However, a distinction between
regional strains related to, for example, shear zones or regional
foliation patterns that may have controlled granite emplacement,
and emplacement-related strains caused by the granitoids themselves, such as granite ballooning and the displacement of wall
rocks, is commonly difficult (Cruden 1998). In both cases, the
superimposition of regional and intrusion-induced strains is
common, and granite emplacement is, in most cases, achieved
through multiple mechanisms that can be both of a regional and
a more local nature (Paterson & Fowler 1993).
By far the most prolific period of granite production and
373
374
Fig. 1. Regional geology of the Barberton
granite–greenstone terrane (after
Anhaeusser et al. 1981) and its location in
the Kaapvaal Craton in southern Africa
(inset).
anorogenic emplacement of the granitoids (e.g. Anhaeusser &
Robb 1983). This interpretation has not remained unchallenged,
and Robb et al. (1983) and Jackson & Robertson (1983)
described the presence of regional-scale gneiss belts within and
along the margins of the batholiths. The multiphase intrusive
relationships between basement gneisses and the GMS suite and
deformation of the potassic granitoids suggests that the emplacement of the 3.1 Ga granitoids is, at least partly, structurally
controlled. As a result of these contrasting views on the contact
relationships and the lack of detailed structural work on the large
batholiths, the emplacement and tectonic setting of the cratonwide plutonic suite have remained somewhat enigmatic.
The present study centres around an area of c. 40 km 3 5 km
along the western and northern margin of the Mesoarchaean, c.
3105 Ma Mpuluzi batholith, one of the most extensively studied
plutons of the GMS suite (Anhaeusser et al. 1981; Anhaeusser &
Robb 1983; Kamo & Davis 1994; Yearron 2003) (Figs 1 and 2).
The aim of this study is to constrain the emplacement mechanisms and magmatic assembly of this large batholith that combines a number of internal and external structural features that
seem typical of many of the GMS suite plutons (Robb et al.
1983). This margin, in particular, discloses highly varying
contact relationships between the younger GMS suite rocks and
basement gneisses that closely reflect the existing controversy
about the syn- v. post-tectonic timing and controls of the granite
emplacement (Anhaeusser & Robb 1983; Jackson & Robertson
1983). Mapping was undertaken on the basis of aerial photographs at a scale of between 1:6000 and 1:10 000, and angular
and spatial distortions were corrected by global positioning
system (GPS) readings. The field-based studies were supplemented by thin-section petrography and whole-rock geochemistry to
characterize different intrusive phases. In addition, geochronolo-
sequences (e.g. De Wit et al. 1987; Armstrong et al. 1990); or
(3) questioning the magmatic models for large parts of the
present-day granite–greenstone contacts altogether, as structurally reworked and subsequently exhumed basement gneisses
(e.g. Dziggel et al. 2002; Kisters et al. 2003).
This study focuses on laterally extensive granite plutons of a
subsequent magmatic episode associated with the intrusion of
vast amounts of granodiorites, monzogranites and syenites, the
GMS suite, at c. 3.1 Ga. Rocks of the GMS suite are found not
only in the Barberton granite–greenstone terrane, but also over
large parts of the Kaapvaal Craton, and their emplacement
coincides with the first stabilization of the central parts of the
craton (De Wit et al. 1992; Kamo & Davis 1994; Poujol &
Anhaeusser 2001). The GMS suite in the Barberton granite–
greenstone terrane shows very different internal and external
characteristics from the earlier TTG suite. Individual plutons
may cover several thousand square kilometres and these composite granitoid bodies have traditionally been referred to as
batholiths, alluding to their compositionally and texturally heterogeneous nature and enormous areal extent (e.g. Anhaeusser et
al. 1981). For the most part, the plutons appear undeformed,
intrusion-related wall-rock strains are only locally recorded, and
intrusive relationships with wall rocks are commonly sharply
discordant (e.g. Hunter 1973; Anhaeusser & Robb 1983; Robb et
al. 1983). Regional studies have demonstrated that most of these
granitoids represent subhorizontal, sheet-like intrusions. The
tabular granites are commonly underlain by so-called migmatite
terranes and dyke complexes that have tentatively been interpreted as the feeders to the overlying granite sheets (e.g. Hunter
1957, 1973; Anhaeusser et al. 1981; Anhaeusser & Robb 1983;
Robb et al. 1983). The sum of these features has traditionally
been interpreted to indicate a ‘passive’, post-tectonic and
Fig. 2. Geological map of the granite–
gneiss terrane south of the Barberton
greenstone belt illustrating the spatial
distribution of the GMS suite and older
TTG gneisses and enclosed greenstone
remnants.
375
The petrographic and geochemical details of the GMS suite have
been given by Anhaeusser & Robb (1983), Anhaeusser (1980)
and Yearron (2003). Intrusive relationships and the salient
petrographic characteristics of the GMS suite are listed in
Distribution of the GMS suite in the study area
of the GMS suite in an extensional and rift-type tectonic setting
was proposed by Kamo & Davis (1994), based on the alkaline
nature of the rocks and the emplacement of some smaller plutons
as NW–SE-trending, distinctly dyke-like bodies (Figs 1 and 2).
Hunter (1957, 1973) was probably the first to establish the
subhorizontal, sheet-like geometry of the Mpuluzi batholith. He
also estimated a thickness of the granitoid sheet of c. 700–
1000 m based on his mapping of the Archaean granitoids in the
mountaineous terrain of Swaziland. The tabular geometry has
since been confirmed in regional field studies by Anhaeusser
(1980) and Anhaeusser & Robb (1983), who also suggested a
very shallow crustal level of emplacement for the Mpuluzi
batholith mainly on the grounds of textural evidence in the
granitoids. The lower- to sub-greenschist-facies metamorphic
conditions of the Barberton greenstone belt to the immediate
north render such shallow emplacement levels likely. However,
there are, as yet, no direct and reliable P–T data that could
constrain the emplacement depth.
The Mpuluzi batholith intrudes into older, c. 3.2–3.5 Ga,
amphibolite-facies, steeply dipping, banded TTG gneisses and
enclosed supracrustal greenstone remnants. Basement gneisses
are parallel to the western, strongly gneissose margin of the
Mpuluzi batholith (Figs 2 and 3) and structural evidence points
to the rotation of the wall-rock gneissosities into parallelism with
this western margin (see below). Notably, a similar belt of
subvertical, NE–SW-trending gneisses within and adjacent to the
Mpuluzi batholith has been described by Jackson & Robertson
(1983) some 30 km SE of the present study area (Fig. 1). Here,
the granites of the Mpuluzi batholith have intruded the southernmost parts of the Barberton greenstone belt, the Motjane schist
belt, and both greenstones and granites have been coaxially
deformed. TTG gneisses and greenstones along the northern
contact of the Mpuluzi batholith are, in contrast, commonly
sharply truncated by the intrusive granites (Fig. 2). The roof
rocks of the Mpuluzi batholith are nowhere exposed.
gical results are presented on older TTG gneisses and younger
potassic intrusive rocks to provide absolute age constraints on
the timing of the emplacement and fabric development in different igneous phases.
The GMS suite in the study area
Rocks of the Mesoarchaean GMS suite in the area studied here
include three main igneous units, namely the Mpuluzi batholith
(sensu lato) and the smaller intrusions of the Boesmanskop and
Weergevonden syenogranites situated along the NW margin of
the Mpuluzi batholith (Fig. 2). The Mpuluzi batholith is a
composite pluton, made up of a number of petrographically and
texturally distinct phases that range in composition from granodiorite, monzonite and monzogranite to syenogranite
(Anhaeusser & Robb 1983; Robb et al. 1983; Yearron 2003).
The semicircular granitoid covers an area of at least 4000 km2
south of the Barberton greenstone belt (Fig. 1). It occupies the
high-lying peneplain between South Africa and Swaziland, and
borders against the low-lying, older TTG–greenstone terrane in
the north along a prominent, 500–700 m high escarpment. Its
southwestern extent is concealed by younger Karoo-aged cover
rocks. Other large batholiths of the GMS suite in the region
include the Nelspruit batholith to the north of the Barberton
greenstone belt and the Heerenveen and Piggs Peak batholith in
the south and SE of the greenstone belt, respectively (Anhaeusser
et al. 1981) (Fig. 1).
U–Pb age constraints from zircons from a fine-grained
granodioritic phase indicate an age of crystallization of
3105 3 Ma for the Mpuluzi granite, whereas the main, coarsegrained phase of the Boesmanskop syenogranite has been dated
at 3107 þ4/2 Ma (Kamo & Davis 1994). These ages are,
within error, identical to the crystallization ages of the large
Nelspruit batholith (3106 3 Ma), and all available age data for
the GMS suite around the Barberton granite–greenstone terrane
suggest a very narrow age range for the emplacement of the
potassic granitoids (Kamo & Davis 1994). The c. 3.1 Ga age of
emplacement of the GMS suite coincides with the regional D3
phase of tectonism described from the northern parts of the
Barberton greenstone belt (De Ronde & De Wit 1994). De
Ronde & De Wit (1994) envisaged a transtensional tectonic
environment for the D3 tectonism. A synkinematic emplacement
376
N
a
Western Domain
WSZ
N
b
N
WSZ
2
4
Central Domain
0
6
10 km
Fig. 3. Structural map of the granite–
greenstone terrane south of the Barberton
greenstone belt. Lower hemisphere, equalarea projections represent poles to the
gneissosity (þ) and mineral stretching
lineations for the northern (inset a) and
southern (inset b) parts of the Welverdiend
shear zone (WSZ). The width of the
Welverdiend shear zone is indicated by the
presence of pervasive solid-state
gneissosities.
content. They show medium- to coarse-grained cumulate-like
textures made up of millimetre-sized, euhedral K-feldspar crystals with interstitial hornblende and/or biotite. Euhedral, millimetre-sized crystals of titanite are locally abundant. Similar
rocks in the region are found only in the dyke-like Kees Zyn
Doorns syenite, some 8 km to the NW (Fig. 1). The southern
margin of the Boesmanskop syenogranite is deformed. This
margin contains a strong NE–SW-trending solid-state gneissosity
Foliations
Lineations
Shear-sense indicators
3.1 GMS suite
3.2 - 3.4 TTG Gneisses
Greenstone lithologies
Legend
Northern Domain
8
Table 1. Anhaeusser & Robb (1983) used the term Boesmanskop
Syenogranite Complex to describe a suite of mineralogically and
texturally distinct rocks that range in composition from monzogranite to syenite. The main outcrops of this rock suite underlie
the two steep-sided hills of the Boesmanskop to the immediate
north of the main escarpment. The rocks at this locality are
syenites, quartz syenites and syenogranites, and are typically
reddish to pinkish in colour owing to their high K-feldspar
Main rock units
Petrography and appearance
Lit-par-lit intrusive relationships
with basement and syenogranite
gneisses (1) along the Welverdiend
shear zone; large, homogeneous
body in the west (Fig. 2)
Boesmanskop syenogranite and
western margin of the Mpuluzi
batholith; contacts with basement
gneisses suggest, at least in parts,
a subhorizontal sheet-like geometry
Mainly in the south in central
domain
As NE–SW-trending dykes
throughout the Welverdiend shear
zone
Intrusive into the northern strike
extent of the Welverdiend shear
zone
Evidence for age
In places protomylonitic; xenoliths occur in
most other phases of the Mpuluzi batholith,
i.e. intruded by (2) and subsequent phases
Locally cut by (5), contains xenoliths of (1)
and (2); solid-state gneissosity along
Welverdiend shear zone; local magmatic
fabric
Solid-state gneissosity and intruded by (3)
and (5), intrusive into (1). U–Pb zircon age
of 3113 2.4 Ma (this study)
No direct intrusive relationships with other
phases of the GMS, locally lineated
Solid-state mylonitic fabrics, intrusive into
(1), (2) and (3), no clear relationship with
(4)
Solid-state gneissosity, intrusive into (1), (2)
and (3), no clear relationship with (5)
Undeformed leucogranites and
Contains enclaves of (1), (5) and (7)
large pegmatite bodies. Dominant
phase along the northern margin
In two localities along the
Weak solid-state gneissosity; intrudes
Welverdiend shear zone, but mainly (1)–(5)
in central domain
Weergevonden tail
Subhorizontal sheets in the interior Undeformed, crosscuts phases related to (8)
of the pluton
Occurrence
Table 1. Summary of the main rock types of the GMS suite, their occurrence and relative age relationships in the study area
(9) Fourth leucogranite
(7) Second leucogranite
(8) Third leucogranite
Medium grey, very fine-grained, K-feldspar,
plagioclase, quartz, minor hornblende,
biotite and muscovite
Pinkish white; fine- to medium-grained;
K-feldspar, plagioclase, quartz, minor
biotite and muscovite
Light grey, fine-grained; K-feldspar,
plagioclase, quartz, minor hornblende, biotite
and muscovite
Light grey, fine- to medium-grained;
K-feldspar, quartz, plagioclase
Medium to dark grey, fine-grained; biotite,
K-feldspar, plagioclase, quartz
(6) Weergevonden
syenogranite
(5) Granodiorite dykes
(4) Augengneiss dykes
(3) Megacrystic phase
(2) First leucogranite
Dark to medium grey with elongated,
whitish pink K-feldspar augen (up to 3 cm);
biotite–hornblende–feldspar–quartz
groundmass
Light grey to pinkish, medium- to coarsegrained, K-feldspar megacrysts (up to 5 cm),
K-feldspar, plagioclase, quartz, muscovite
groundmass
Light grey to pinkish grey, medium- to
coarse-grained; K-feldspar, plagioclase,
quartz, minor hornblende, biotite and
muscovite
(1) Boesmanskop syenites Reddish to pinkish, K-feldspar, hornblende,
and syenogranites
biotite, minor titanite, quartz, plagioclase;
considerable textural and mineralogical
variations
377
The Mpuluzi batholith is bounded in the west by subvertical,
NE–ENE-trending gneisses that show widespread protomylonitic
textures. The high-strain fabrics are pervasively developed in
both basement gneisses and greenstones as well as intrusive
rocks of the younger GMS suite. Non-coaxial shear fabrics are
common and this western gneiss belt is referred to as the
Welverdiend shear zone (Fig. 3), based on the farm Welverdiend
where shear fabrics are best developed.
The Welverdiend shear zone has an arcuate trend from NE in
the south to more ENE along its northern extent (Fig. 3a). The
shear zone can be traced for c. 25–30 km along strike and
the presence of subvertical, gneissose fabrics in wall rocks and
the Mpuluzi granite suggests a width of c. 3–5 km. Banded TTG
gneisses to the west of the Welverdiend shear zone show
predominantly moderate dips (30–408), but progressively rotate
and steepen into parallelism with the subvertical shear fabrics
over a distance of 300–500 m (Figs 3 and 4a). Metre-scale
mushroom-type interference folds are contained in the steep
fabric of the Welverdiend shear zone, suggesting the pervasive
refoliation and refolding of earlier fabrics and folds contained in
the TTG gneisses by the shear zone. The southern extent of the
shear zone is covered by younger Karoo strata. Its northern,
rather abrupt termination is marked by the NW–SE-trending
Weergevonden syenogranite, beyond which there is no evidence
of the ENE-trending shear fabrics. Mineral stretching lineations
are defined by elongated quartz and quartz–feldspar mineral
aggregates as well as stretched biotite clots and are locally well
developed. The lineations show shallow easterly plunges in the
north becoming steeper in the south of the Welverdiend shear
zone (Fig. 3b). Most granitic rocks along the Welverdiend shear
zone show pervasive solid-state fabrics evidenced by the dynamic
recrystallization of all mineral components (Fig. 5b). Mafic
minerals such as amphibole and/or biotite form part of the
protomylonitic fabrics developed in greenstones and granitoids
and appear largely unaltered without signs of retrogression.
These features point to deformation under amphibolite-facies
conditions. Retrograde brittle–ductile shearing is locally indicated by minor chloritization and epidotization along narrow,
foliation-parallel cataclastic zones.
Shear-sense indicators are abundant along the northern extent
of the Welverdiend shear zone. For example, large pavements
along the southern, gneissose margin of the Boesmanskop
syenogranite are entirely made up of closely spaced S–C fabrics
Western domain: the Welverdiend shear zone
underlain by massive and largely undeformed fine-grained
leucogranite in the north grading into coarsely porphyritic granite
in the south (Anhaeusser & Robb 1983). Textural variations
between outcrops are common and point to the rather heterogeneous nature of the granite. Granodiorites and porphyritic
granites of the ‘bimodal association’ (Anhaeusser & Robb 1983)
typically show irregular, interfingering intrusive relationships
with, in places, diffuse and gradational contacts. The predominant megacrystic phase of the Mpuluzi granite locally preserves
magmatic fabrics defined by the alignment of euhedral, commonly zoned K-feldspar laths, but with little evidence of
regionally consistent trends. Pegmatite and granodioritic dykes
form stockworks or irregularly shaped bodies. Towards the west,
the feldspar megacrysts show a preferred orientation defining a
NE-trending magmatic fabric. This fabric is progressively overprinted by a pervasive high-temperature gneissosity defined by
feldspar augen and quartz ribbons approaching the western
domain (Figs 3 and 5a).
(Fig. 3) and the gradual strain increase from undeformed
syenogranites in the NW to gneissic rocks in the SE can be
followed over a distance of c. 500 m. Our regional mapping
shows large tracts along the escarpment to the south and SW of
the Boesmanskop pluton to be made up of similar pink gneisses.
These gneisses are connected to the main outcrop of the
Boesmanskop pluton, resulting in a different outcrop pattern of
the syenogranites compared with that shown on published maps
(Anhaeusser et al. 1981) (Fig. 2). Xenoliths of the pink syenitic
gneisses are found in almost every other phase of the Mpuluzi
batholith, indicating that the syenites form one of the earliest
phases of the GMS suite in the region.
The Weergevonden syenogranite is a NW–SE-trending dykelike intrusion that measures c. 8 km 3 1 km (Anhaeusser 1980)
(Figs 1 and 2). The syenogranite, also referred to as the
Weergevonden tail (Anhaeusser et al. 1983), is a leucocratic,
light grey and fine- to medium-grained rock, and is texturally
and mineralogically distinct from the adjacent Boesmanskop
syenogranite (Anhaeusser 1980). Rocks of the Weergevonden tail
lack a macroscopically visible foliation but contain, in places, a
steep northerly plunging lineation. The dyke-like syenogranites
sharply truncate the ENE-trending gneissosity developed in, for
example, gneisses related to the Boesmanskop syenogranite and
the older TTG gneisses and greenstones (see below).
The main body of the Mpuluzi batholith is made up of three
major phases, including: (1) a variety of fine- to medium-grained
leucogranites; (2) coarsely porphyritic and often megacrystic
granites; (3) locally developed fine- to medium-grained, dark
grey granodiorites (Anhaeusser & Robb 1983) (Fig. 2). Pegmatite
dykes, stockworks or large, irregularly shaped pegmatite pods are
ubiquitous and particularly abundant along the western margin
and on the topographically high-lying areas in the central parts
of the Mpuluzi batholith. The northern and western parts of the
study area are dominated by a variety of fine- to medium-grained,
light grey to pink–grey leucogranite bodies. The leucogranites
are mainly composed of microcline, plagioclase and quartz with
only minor amounts of biotite and hornblende. Leucogranites in
the western parts of the Mpuluzi batholith contain almost
invariably a subvertical, NE-trending solid-state gneissosity
defined by flattened quartz grains and the grain-shape preferred
orientation of recrystallized feldspar aggregates. The northern
parts of the Mpuluzi batholith, in contrast, are underlain by finegrained leucogranites that appear undeformed in outcrop. Xenoliths of gneissose leucogranites within the undeformed leucogranite together with intrusive relationships point to the sequential
emplacement of the leucogranite phases. The southern, topographically highest parts of the Mpuluzi batholith are made up of
coarsely porphyritic monzogranite characterized by abundant
microcline megacrysts. Fine- to medium-grained, dark grey
granodiorites are locally intrusive into the megacrystic granite,
forming a common spatial association termed the ‘bimodal
association’ by Anhaeusser & Robb (1983).
Structural domains
The NW parts of the Mpuluzi batholith studied here are
subdivided into a central, western and northern domain each
characterized by distinctly different strains and intrusive relationships (Figs 3 and 4).
Central domain
The central domain encompasses the high portions of the
Mpuluzi granite on top of the escarpment. This domain is mainly
378
Fig. 4. Schematic cross-sections taken across the western domain (a) and northern domain (b) illustrating different intrusive relationships between rocks
of the GMS suite and wall rocks (see text for detailed discussion).
Fig. 5. (a) K-feldspar megacrysts of the Mpuluzi batholith defining a strong NE–SW-trending fabric parallel to the western gneissose margin of the
Mpuluzi batholith. At this locality (26818.959S, 30856.009E), magmatic fabrics defined by the alignment of euhedral and undeformed megacrysts are
progressively overprinted by a high-T solid-state gneissosity approaching the western domain. The high-T, solid-state origin of this gneissosity is evidenced
by the marginal recrystallization of megacrysts, pervasive recrystallization of the finer-grained groundmass and quartz ribbons. The top part of the
photograph is made up of an intrusive leucogranite dyke that also contains a solid-state gneissosity. (b) Solid-state, protomylonitic gneissosity in coarsegrained (right-hand side of photo) and medium-grained (top left corner) variety of the Boesmanskop syenogranite (oblique plan view; length of pen is c.
15 cm). The K-feldspar megacrysts are marginally and/or pervasively recrystallized to form an augen texture and mafic minerals (biotite and hornblende)
are unretrogressed. The deformation textures and mineral assemblages testify to the high-T origin of the protomylonitic fabric. Locality: 26809.509S,
30868.959E (Mhlingase river, SE of the Boesmanskop). (c) S–C fabric relationships in syenitic gneiss indicating dextral sense of shear, southern margin of
the Boesmanskop syenogranite (26806.059S, 30869.179E). (d) Late-stage, cross-cutting and openly folded pegmatite dyke intruding into the Welverdiend
shear zone (shear fabrics of the Welverdiend shear zone run approximately horizontal in the photograph). Fold axes trend NE–SW, parallel to the
Welverdiend shear zone, and folding indicates a bulk NW–SE-directed shortening at high angles to the trend of the Welverdiend shear zone. Locality:
26817.729S, 30857.289E (east of the Schapenburg schist belt). (e) Lit-par-lit intrusive relationships between tonalitic basement gneiss (dark grey) and
leucogranite veins (light grey). (Note the mylonitic fabrics and augen textures developed in the leucogranite veins.) Locality: 26813.889S, 30858.629E
(foothills of the western escarpment). (f) Tightly folded aplite vein in syenitic gneiss. The gneissosity in the syenitic gneiss (annotated, S) is axial planar to
the fold. A leucogranite dyke is intrusive into the syenitic gneiss on the right-hand side of the photograph. The leucogranite is itself strongly gneissose.
Locality: 26810.209S, 30862.009E (SE margin of the Boesmanskop syenogranite). (g) Mosaic-like intrusive breccia of leucogranite (light grey) into a
melanocratic variety of the syenogranite gneisses related to the Boesmanskop intrusion (dark grey); lens cap in upper central parts of photo for scale.
Locality: 26812.129S, 30861.509E (east of the Boesmanskop syenogranite). (h) Intrusive breccia of weakly foliated leucogranite (light grey) containing
angular fragments of foliated granodiorite (dark grey). A solid-state foliation in the leucogranite runs from the lower left-hand to the upper right-hand
corner of the photograph. Locality: 26812.839S, 30861.129E (central parts of the Welverdiend shear zone).
379
380
The generally gneissose, NW-trending margin of the Mpuluzi
batholith of the western domain shows a characteristic swing to
more easterly trends at and close to the termination of the
Welverdiend shear zone (Fig. 3). Beyond the shear-zone termina-
Northern domain
structurally higher levels in the Mpuluzi granite. Within c. 1 km
distance from the Welverdiend shear zone, granodiorite dykes
commonly contain a pervasive, dyke-parallel solid-state foliation
with locally abundant tight to isoclinal intrafolial folds defined
by thin aplite veins. Rodding fabrics or lineated augen textures
are also developed. In general, the strain intensity is commonly
considerably higher in the granodiorite dykes than in the tonalitic
wall-rock gneisses, which probably reflects strain localization
into the dykes (e.g. Zulauf & Helferich 1997). Deeply incised
river sections offer exposures east of and away from the
Welverdiend shear zone into the underlying levels of the main
mass of the Mpuluzi batholith. Here, the dykes may still contain
a subvertical, NE-trending solid-state gneissosity, but show clear
evidence of a magmatic foliation defined by clusters of aligned,
elongated microgranitic and/or mafic enclaves. The trains and
clusters of enclaves are parallel to the dyke walls and the solidstate gneissosity that characterizes the western margin of the
Mpuluzi batholith.
Multiple sheeting and dyke-in-dyke intrusive relationships of
the GMS suite rocks are common along the entire strike extent
of the Welverdiend shear zone. Angular xenoliths of, for
example, leucogranite gneisses in granodiorite dykes or granodioritic fragments in later leucogranites (Fig. 5g and h) commonly contain solid-state gneissosities. This demonstrates the
emplacement of the GMS suite over a protracted period of time,
into fully crystallized earlier intrusive rocks of the same rock
suite and during regional deformation. Both the fabric orientation
and fabric intensity in the intrusive dykes indicate that this
deformation is related to shearing along the Welverdiend shear
zone. Chilled margins are absent, suggesting the intrusion of the
sheets into hot wall rocks, consistent with the inferred high-T
conditions of deformation.
Subhorizontal or shallowly dipping granite sheets are relatively
rare along the Welverdiend shear zone. Tightly folded dykes
occur as well as undeformed and highly discordant sheets, which
suggests the syn- to post-kinematic timing of their emplacement.
Most of these sheets vary in width between c. 1 and 10 m.
Contact relationships as well as the regional outcrop pattern
suggest that the pinkish syeno-granites and syenites of the
Boesmanskop syenogranite form, at least in parts, subhorizontal
intrusive sheets in the western domain adjacent to the Welverdiend shear zone. The contact between the syenogranites and
basement gneisses is well exposed in the river bed of the
Theespruit River to the immediate NW of the escarpment and
south of the Boesmanskop (Fig. 6a and b). This contact is sharp
and forms a subhorizontal, slightly undulating plane along which
the syenogranites sharply truncate the subvertical, banded TTG
gneisses and amphibolite-facies greenstones. Dykes and veinlets
of syenogranite are contained in or cross-cut the gneissosity of
the underlying TTG gneisses at low angles. In numerous places
the dykes can be seen to be connected to the overlying sheet-like
syenogranites (Fig. 6b). The dykes and veinlets are commonly
folded in the foliation of the enveloping gneisses and greenstones, and contain a variably developed solid-state foliation,
indicating their synkinematic emplacement during deformation
of the basement gneisses. The overlying syenogranites, however,
appear massive and undeformed in outcrop.
that are pervasively developed over several tens of metres. Mica
fish, and rotated ó- and ä-clasts are also present, and shear-sense
indicators consistently point to a dextral sense of shear (Fig. 5c),
corresponding to the shallow easterly plunge of the mineral
stretching lineation. Non-coaxial shear fabrics and kinematic
indicators are, in contrast, scarce along the NE-trending, southern
extent of the Welverdiend shear zone. Pegmatite and leucogranite
dykes that cross-cut the shear zone at high angles are openly to
tightly folded into upright, symmetrical folds (Fig. 5d) and fold
transposition, which is widespread along the northern extent of
the Welverdiend shear zone, is rare. Granitoid sheets that have
intruded subparallel to the foliation commonly show chocolatetablet type boudinage. The fold geometries and chocolate-tablet
boudinage of intrusive dykes point to a large component of
NW–SE-directed subhorizontal bulk shortening perpendicular to
the foliation in this southern part of the Welverdiend shear zone.
Intrusive relationships. The Welverdiend shear zone is intruded
by a variety of granitoids related to the GMS suite, including
leucogranites, monzogranites, granodiorites, syenogranites and
quartz syenites together with abundant aplites and granite
pegmatites (Table 1). Most of the granitoids form subvertical
sheets that are concordant with the subvertical gneissosity in the
Welverdiend shear zone; that is, they are sills or foliation-parallel
and -subparallel dykes (Figs 4a and 5e, f). The subvertical
granite sheets vary in width from centimetres to several tens of
metres and show strike lengths of several hundred metres to
kilometres. However, subhorizontal and sharply discordant sheets
also occur (Fig. 4a). The subhorizontal sheets are relatively rare
in the foothills of the escarpment, but become more common at
higher structural levels.
Subvertical granite sheets intrude in a lit-par-lit manner (Figs
4a and 5e), parallel or at low angles to the gneissose fabric of the
Welverdiend shear zone. In contrast to the assertion of
Anhaeusser & Robb (1983) that the gneissose fabrics in the
GMS suite represent an old fabric inherited from subsequently
K-metasomatized TTG gneisses, cross-cutting relationships and
the different degrees of post-emplacement deformation indicate
emplacement of the various phases of the GMS suite during
progressive deformation along the Welverdiend shear zone. Earlier sheets of foliation-parallel leucogranites and pegmatites are
pervasively mylonitized (Fig. 5e). The intrusive granitoids show
feldspar-augen textures, transposition of fabrics and large quartz
ribbons that are all parallel to the external foliation of the
Welverdiend shear zone. Cross-cutting dykes are folded, partly
transposed or boudinaged in the gneissose foliation (Fig. 5f).
Late-kinematic sheets and dykes may still preserve primary
intrusive relationships such as horn-and-bridge structures, but are
also typically gneissose. Late- to post-kinematic leucogranites
and pegmatites are sharply discordant and cross-cut all earlier
intrusions and shear-zone fabrics, forming areally extensive netveined or stockwork-like intrusive breccias, particularly at higher
levels and on top of the escarpment (Figs 4a and 5g, h).
Medium to dark grey, fine- to medium-grained granodiorites
form a distinct set of subvertical NE-trending dykes. Along the
escarpment, these dykes can be seen to structurally underlie the
main, sheet-like Mpuluzi granite. The width of the dykes ranges
from several metres to tens of metres and individual dykes can
be followed vertically and along their NE strike for several
hundred metres, forming a kilometre-scale anastomosing network
along the eastern margin of the Welverdiend shear zone.
K-feldspar, plagioclase, quartz and biotite are the main
rock-forming minerals and the dykes are mineralogically and
texturally similar to the fine-grained granodiorites found at
381
there may be far more dykes related to the GMS suite in this
region. At higher structural levels, subhorizontal sheet-like
leucogranites and pegmatites become more abundant. Stockwork-like intrusive breccias result from the intersection and
linkage of subhorizontal sheets with steeply dipping dykes (Fig.
4b). The dykes and sheets sharply truncate structures in the wall
rock gneisses and greenstones, and large (several tens of metres)
wall-rock xenoliths may be completely engulfed by the intrusive
sheets. The orientation of the gneissosity in the wall-rock
xenoliths commonly suggests no or very little rotation of the
xenoliths with respect to the undisturbed TTG gneisses at the
base of the escarpment. The continuity of structural trends from
basement gneisses to large-scale basement xenoliths contained
within the Mpuluzi granite preserves a ‘ghost stratigraphy’ (e.g.
Pitcher 1970; Hutton 1992). A notable difference between dykes
and sheets at lower structural levels and those higher up is that
the latter appear undeformed. The topographically higher parts
of the Mpuluzi batholith are made up of mainly massive, finegrained leucogranite intermingled with irregular pods and dykes
Fig. 6. (a) The subhorizontal contact between the syenites of the Boesmanskop pluton sharply truncating basement gneisses and amphibolites (bottom) in
the Theespruit River valley to the immediate south of the Boesmanskop pluton (26805.929S, 30866.289E). (Note the undeformed dyke attached to the
sheet-like syenites.) Cross-sectional view (looking SW), field of view is c. 1.5 m across. (b) Plan view of the subhorizontal eroded contact between
syenites of the Boesmanskop pluton (top of the photograph) and underlying subvertical amphibolites (black) and minor TTG gneisses (dark grey) (same
locality as (a)). The patches of syenite overlie subvertical amphibolites along a sharp subhorizontal plane. The subhorizontal sheet is connected to
subvertical sheets contained within the amphibolites (bottom half of photograph). Some of the subvertical sheets are folded within the foliation of the
basement (below lens cap). (c) Granitic dyke (light grey, centre of photograph) cross-cutting banded TTG basement gneisses in the foothills of the
northern escarpment (26806.199S; 30849.819E). The dyke contains isolated K-feldspar megacrysts. (d) Shallowly dipping sheet of fine-grained leucogranite
(central parts of the cliff) intrusive into late-stage pegmatites on top of the escarpment in the structurally higher portions of the Mpuluzi batholith in the
central domain (26812.539S; 30868.179E). The cliff is c. 8 m high.
tion, the granites no longer intrude as mainly concordant or
subconcordant sheets parallel to the Welverdiend shear zone, but
rather as highly discordant dykes that cut at variable angles
across the structural trend of the older TTG basement gneisses
and greenstones. On a regional scale, this northern margin
appears as a transition up to several kilometres wide from
isolated granite dykes intrusive into TTG basement gneisses in
the north, through stockwork-like intrusion breccias into the
massive Mpuluzi granite of the central domain (Fig. 4b). Moreover, rocks related to the Mpuluzi batholith appear, for the most
part, undeformed.
The lower levels in the northern foothills of the escarpment
contain dykes of leucogranite, porphyritic granite and minor
pegmatites that intrude the TTG basement gneisses (Figs 4b and
6c). The commonly discordant, subvertical granite dykes show
scattered trends, but ESE-trending dykes predominate. However,
most intrusive dykes contain a dyke-parallel gneissosity, which,
together with the compositional similarities to the TTG basement
gneisses, often complicates their recognition, and we suspect that
382
Sample 586 represents medium-grained, greyish–pinkish leucogranite related to the Mpuluzi batholith. The granite consists of
K-feldspar, plagioclase and quartz, with only minor amounts of
biotite and hornblende. A gneissosity is defined by elongated
quartz grains and quartz–feldspar aggregates. Zircons were
prismatic, pink to reddish in colour and translucent. As for
sample 588b, CL imaging of the grains is characterized by
concentric magmatic zoning. Few grains show what appears to
be a core surrounded by an overgrowth. Seven grains were
analysed (Table 2) and plotted in a concordia diagram (Fig. 7c).
Five of the seven grains (Zr1, Zr2, Zr3, Zr4 and Zr7) define an
upper intercept age of 3113.2 2.4 Ma. This age is suggested to
represent the time of crystallization of the granite. Two points
(Zr5 and Zr6) plot below this discordia in a very discordant
position. Their position could be the consequence of the presence
of core and overgrowth. Previous single zircon ages from
undeformed potions of the GMS suite have pointed to a very
narrow age range of 3105 3 Ma for the entire suite (Kamo &
Davis 1994). The age of the leucogranite obtained in this study
is, thus, the first indication that the emplacement of the GMS
Sample 586: Mpuluzi granite
plot in slightly discordant to very discordant positions and do not
define a simple, single group or trend, which may indicate the
effects of more than one Pb-loss event. Nevertheless, we interpret
this complex age pattern as follows. Four grains (Zr 1, 3, 4 and
7, Fig. 7a), define an upper intercept age of 3228 12 Ma that
we consider as representative of the emplacement age of this
tonalite. The three remaining grains (Zr2, Zr5 and Zr6, Fig. 7a)
plot above the discordia defined by the other zircons and could
therefore reflect a complex Pb-loss caused by a metamorphic
event and recent Pb-loss. The U–Pb zircon age of 3228 12 Ma
is, within error, identical to the 3231 5 Ma age obtained for a
gneissose tonalite some 5 km to the south in the Schapenburg
schist belt (Stevens et al. 2002) as well as the large Kaap Valley
tonalite in the north of the Barberton greenstone belt (Kamo &
Davis 1994). The present mapping has also shown that the two
tonalite bodies contained in the Welverdiend shear zone are
probably part of a single, more extensive tonalite pluton in the
southern granite–gneiss terrane (Fig. 2).
The tonalitic gneiss 588b contains a pervasive subvertical, NEtrending foliation parallel to the foliation of the Welverdiend
shear zone. This raises the question of the timing of the fabricforming event, as intrusive and structural relationships along the
Welverdiend shear zone point to the c. 3.1 Ga intrusion of the
GMS suite during high-temperature deformation. The gneissosity
in the tonalite is defined by aligned hornblende and/or biotite
and flattened quartz–feldspar aggregates. Titanite forms a relatively abundant accessory mineral aligned in the foliation. Five
multi-fragment fractions of these titanites were analysed and data
are reported in Table 2. Plotted in a concordia diagram (Fig. 7b),
they plot in a concordant position, except for Sph5, and define
an upper intercept age of 3124.1 1.6 Ma. The four concordant
titanite fractions analysed define a concordia age of
3121.9 1.1 Ma that we consider as the age of crystallization of
the titanite. Titanite has a high closure temperature for its U–Pb
system (c. 630–730 8C; see Frost et al. 2000, for references). As
the measured age of a mineral represents the age when the
mineral passed through its closure temperature, this age of
3.12 Ga can be interpreted as the age of the peak of the
amphibolite-facies metamorphic event and ductile deformation
that affected this tonalitic gneiss.
of pegmatites (Fig. 4b). However, the sheeted nature of the
Mpuluzi granite is locally evidenced by subhorizontal, finegrained leucogranite sheets that intrude and sharply truncate even
late-stage pegmatites (Fig. 6d). Where exposed, both the hanging-wall and footwall contacts of the sheets are sharp. The
thickness of the sheets ranges from c. 1.5 m (the thinnest sheets
observed where footwall and hanging-wall contacts are exposed)
to probably well in excess of 10 m. Xenoliths of TTG gneisses
and greenstones are not as common as at lower structural levels
to the north and commonly have a random orientation suggesting
some degree of rotation (Fig. 4b). Xenoliths of earlier phases of
the GMS suite include medium-grained leucogranite gneisses
that form large bodies along the Welverdiend shear zone, and
pink syenitic gneisses related to the Boesmanskop pluton.
Geochronology
U–Pb zircon and titanite ages were obtained from a tonalitic
basement gneiss along the eastern margin of the Welverdiend
shear zone (sample 588b) and a weakly foliated leucogranite
related to the GMS suite (sample 586). These samples were
analysed to confirm the c. 3.1 Ga age of deformation along this
hitherto unrecognized shear zone and also to potentially provide
estimates of the duration of the tectonism and plutonism.
U–Pb zircon and titanite ID-TIMS technique
Mineral separates were prepared from 4–6 kg rock samples. Rock
samples were pulverized using a heavy-duty hydraulic rock splitter, jaw
crusher and swing mill. Mineral separation involved the use of a Wilfley
Table, heavy liquids (bromoform and methylene iodide) and a Frantz
Isodynamic Separator.
Analyses were performed at Memorial University of Newfoundland,
Canada. Normal transmitted and reflected light microscopy as well as
SEM back-scattered or cathodoluminescence (CL) imagery were used to
determine the zircon internal structures prior to analysis. Handpicked
zircons and titanites were abraded (Krogh 1982) then washed in dilute
nitric acid and ultra-pure acetone. Single grains or small populations of
zircons and titanites were then placed into 0.35 ml Teflon vials together
with HF and few drops of HNO3 and a mixed 205 Pb– 235 U spike. Eight of
these Teflon vials were then placed in a Parr Container for several days at
210 8C (Parrish 1987). The samples were measured on a Finnigan
MAT262 mass spectrometer equipped with an ion-counting secondary
electron multiplier. A detailed account of the entire analytical technique
has been given by Dubé et al. (1996).
Total Pb blanks over the period of the analyses range from 5 to 1 pg
and a value of 5 pg was assigned as the laboratory blank
(206 Pb=204 Pb ¼ 18:97 1, 207 Pb=204 Pb ¼ 15:73 0:5 and 208 Pb=204 Pb ¼
39:19 1:5). The calculation of common Pb was carried out by
subtracting blanks and then assuming that the remaining common Pb has
an Archaean composition determined from the model of Stacey &
Kramers (1975). Data were reduced using PbDat (Ludwig 1993).
Analytical uncertainties in Table 2 are listed at 2ó and age determinations
were processed using Isoplot/Ex (Ludwig 2000).
Sample 588b: tonalitic gneiss
Sample 588b is from a medium-grained tonalitic gneiss taken to
the immediate east of the Welverdiend shear zone. Zircons
extracted from this sample were typically prismatic, red to
yellow–whitish in colour and translucent to opaque. CL imaging
revealed that they are usually concentrically and compositionally
zoned without apparent core and/or rim. Five red translucent and
two white–yellow grains were analysed (Table 2). The Th/U
ratios vary in the range of 0.4–0.6 for the first type and 0.2–0.4
for the second. Plotted in a concordia diagram (Fig. 7a), they
Table 2. Isotope dilution thermal ionization mass spectrometry U–Pb data for samples 588b (zircons and titanites) and 586 (zircons)
383
384
U
Pb/238 U
206
0.65
0.55
0.45
13
6
2100
0.35
0.25
0.66
0.62
0.58
0.54
0.50
0.46
0.6
0.5
0.4
0.3
0.2
0.1
2500
2000
Pb/
2950
17
235
18
U
22
84
494
191
129
120
87
59
57
46
80
60
186
64
186
88
100
80
59
39
55
49
68
51
120
0.6
0.2
0.4
0.5
0.5
0.4
0.5
2.1
2.6
1.3
1.2
0.9
2632
718
1319
3799
1021
4891
875
141
124
114
92
382
0.6156
0.3367
0.3650
0.6375
0.5580
0.5672
0.5493
0.6215
0.6217
0.6196
0.6224
0.4907
0.5
0.2
1.6
0.3
0.3
0.3
0.3
0.3
0.2
0.4
0.4
0.2
21.566
9.368
11.029
22.474
18.821
19.412
18.856
20.589
20.614
20.538
20.614
14.932
0.5
0.2
1.6
0.3
0.3
0.3
0.3
0.3
0.3
0.5
0.5
0.2
0.2541
0.2018
0.2191
0.2557
0.2447
0.2482
0.2490
0.2403
0.2405
0.2404
0.2402
0.2207
0.08
0.08
0.08
0.07
0.06
0.14
0.13
0.11
0.16
0.13
0.16
0.07
3210
2841
2974
3220
3150
3174
3178
3122
3123
3123
3121
2986
3092
1871
2006
3179
2858
2896
2822
3116
3117
3108
3119
2574
0.99
0.92
0.71
0.97
0.98
0.93
0.94
0.93
0.84
0.96
0.94
0.95
3
4
1
4
3
3
3
251
69
22
111
267
140
27
165
47
16
51
45
41
19
2.6
0.5
0.7
0.3
0.1
0.4
0.6
448
879
305
512
540
177
567
0.2675
0.5658
0.5353
0.3961
0.1495
0.2345
0.5840
0.3
0.5
0.6
0.4
0.3
0.3
0.6
6.669
18.319
17.122
11.665
3.957
6.192
19.002
0.6
0.5
0.6
0.4
0.3
0.3
0.6
0.1808
0.2348
0.2320
0.2136
0.1919
0.1915
0.2360
0.48
0.11
0.23
0.10
0.21
0.17
0.25
2660
3085
3066
2933
2759
2755
3093
1528
2891
2764
2151
898
1358
2965
0.46
0.97
0.93
0.96
0.84
0.85
0.91
3114
16
3118
3122
3126
Concordia Age
3121.9 1.1 Ma
Pb/235U
21
20
The commonly observed spatial and temporal relationship between deformation and granite emplacement may be interpreted
to reflect either shear-zone assisted melt transfer or strain
localization related to the injection of magma along shear zones
(e.g. Vauchez et al. 1997; Brown & Solar 1998). A distinction
between the two may not always be possible. In the case of the
Welverdiend shear zone, clues to the timing relationship between
intrusion and deformation are potentially provided by the welldefined titanite ages from the tonalitic gneiss within the Welverdiend shear zone. Assuming that this age of 3124.1 1.6 Ma
represents the age of initial high-temperature deformation along
the Welverdiend shear zone, then shearing has commenced well
before the intrusion of the main phase of GMS magmatism at c.
3105 Ma. Dextral shearing along the Welverdiend shear zone is,
thus, probably a manifestation of a regional deformational event
and melt transport was, at least initially, controlled and assisted
by the deformation. The timing of deformation coincides with
the D3 tectonism described by, for example, De Ronde & De Wit
(1994) from the northern margin of the Barberton greenstone
belt and confirms the contention of Jackson & Robertson (1983)
of synkinematic emplacement of the Mpuluzi batholith during
regional deformation. The available age data also suggest that
deformation along the Welverdiend shear zone may have
occurred over a period of c. 15 to 20 Ma, that is, between
c. 3124.1 1.6 Ma and 3105 3 Ma, the younger age bracket
given by the intrusion of the fine-grained, undeformed granodiorite phase dated by Kamo & Davis (1994). Progressive dextral
strike-slip shearing was then, however, accompanied by the
emplacement and repeated injection of the mainly foliationparallel, concordant sheets of the GMS suite. The U–Pb zircon
age of 3113 2.4 Ma for a leucogranite obtained in this study is
significant in this context. It illustrates episodic magma injection
and the assembly of the Mpuluzi batholith between at least
3113 2.4 Ma and 3105 3 Ma; that is, over a period of at
least 3 Ma and up to 13 Ma. The positive feedback effect
between magma injection and deformation (e.g. Zulauf &
Helferich 1997; Vigneresse & Tikoff 1999) is evidenced by the
partitioning of strain into the intrusive sheets along the Welver-
Synkinematic emplacement of the GMS suite
There are two main features that seem pertinent for an understanding of the emplacement and assembly of the Mpuluzi
batholith and related phases of the GMS suite in the area. (1)
The Mpuluzi batholith is bounded in the west by the synmagmatic, NE-trending dextral transcurrent Welverdiend shear zone.
The termination of the Welverdiend shear zone coincides with a
swing of the margin of the Mpuluzi batholith through c. 608 to
ESE trends. Beyond the shear-zone termination, granites of the
GMS suite are intruded as highly discordant sheets and appear
largely undeformed. (2) The main mass of the Mpuluzi batholith
is essentially made up of granite sheets, and both field and
geochronological evidence point to the repeated and multiple
injection of magma. Subvertical sheets and dykes dominate at
lower structural levels. Higher structural levels record the rapid
transition from subvertical to subhorizontal sheets that build up
the main body of the tabular Mpuluzi batholith. In the following,
we will address these features and their significance for the
emplacement of the GMS suite in more detail.
Discussion and conclusions
suite occurred over a protracted period of time and considerably
longer than previously thought.
3300
b
3150
data-point error ellipses are 2ó
3110
3050
0.627
0.625
0.623
0.621
0.619
0.617
207
0.615
20.35 20.45 20.55 20.65 20.75
19
c
2800
Pb/235U
12
Intercepts at
766 ± 9 [±9.1] & 3113.2 ± 2.4 Ma
MSWD = 1.10
Pb/235U
207
207
2400
8
26
3227.7 ± 7.1 Ma
3100
3101 +37/-30 Ma
2900
a
Zr 5 and Zr 8: Zircons are pink and not very translucent
0.283<208Pb/206Pb<0.306
207
14
Zr 1, 3, 4, 6 and 7: zircons are salmon pink and translucent
0.147<208Pb/206Pb<0.182
2700
Sample 588b
2300
10
Sample 588b
(sphenes)
1600
Sample 586
15
Sph5
2850
Intercepts at
1145 ± 12 & 3124.1 ± 1.6 [±9.5]
Ma
MSWD = 1.16
2750
1200
4
Fig. 7. U–Pb concordia diagrams for: (a) zircons from sample 588b,
tonalitic gneiss; (b) titanites from sample 588b; (c) zircons from sample
586, leucogranite gneiss.
0
Pb/238U
206
U
238
Pb/
206
238
Pb/
206
Pb/238U
206
Corr. coeff.
8
3
10
4
7
9
4
15
13
10
10
4
Sample 588b
Zr 1, Pr, R, T
Zr 2, Pr, YW, D
Zr 3, Pr, YW, D
Zr 4, Pr, R, T
Zr 5, Pr, R, T
Zr 6, Pr, R, T
Zr 7, Pr, P, T
Ti 1, 5 Fgts, R
Ti 2, 4 Fgts, R
Ti 3, 7 Fgts, R
Ti 4, 5 Fgts, R
Ti 5, 1 Fgt, R
Sample 586
Zr 1, Pr, R, T
Zr 2, Pr, R, T
Zr 3, Pr, P, T
Zr 4, Pr, P, T
Zr 5, Pr, P, T
Zr 6, Pr, P, T
Zr 7, Pr, R, T
Pb/206 Pb
207
Pb/206 Pb
207
Pb/235 U
207
Pb/238 U
206
Apparent age (Ma)
Radiogenic ratios
Pb/204 Pb
206
Th/U
Pb
(ppm)
U
(ppm)
Weight
(ìg)
Grain
385
One of the salient features of the Mpuluzi batholith is that it
appears to be constructed of granite sheets. Regional-scale maps
of the southern granite–gneiss terrane depict the close spatial
relationship between granite dyking and sheeting and the perimeters of the Mpuluzi batholith in what Anhaeusser et al. (1981)
mapped as a marginal ‘migmatite belt’ surrounding the granitoid.
The position of the marginal migmatite belt closely corresponds
to the location of the escarpment, thereby exposing the structural
Assembly of the Mpuluzi batholith
recorded by earlier workers (Anhaeusser & Robb 1983; Jackson
& Robertson 1983) as well as the arcuate map pattern of the
exposed northern margin of the Mpuluzi batholith.
Fig. 8. Synoptic sketch of the envisaged emplacement of the Mpuluzi
batholith. (a) Initial dextral transcurrent shearing along the Welverdiend
shear zone (WSZ) and the Motjane schist belt (Jackson & Robertson
1983) during D3 -related NW–SE subhorizontal shortening.
(b) Progressive deformation is accompanied by the emplacement of
subvertical, sheet-like intrusions parallel to the shear zones, and early
subhorizontal sheets such as the Boesmanskop syenogranite. The en
echelon arrangement of the bounding shear zones results in a dilational
jog geometry (inset). Emplacement of subvertical dykes and sheets into
the dilational jog is related to progressive deformation along the
bounding shear zones. (c) Granite sheeting continues during further
deformation. The transition from subvertical dykes to subhorizontal
sheets at the ‘critical depth’ results in the assembly of the multiphase,
tabular Mpuluzi batholith. The internal assembly of the batholith is via
granite sheeting; external controls are provided by regional transcurrent
shearing and associated dilation.
diend shear zone. Continued and repeated magma injection of
the foliation-parallel sheets has probably resulted in higher strain
rates along the Welverdiend shear zone because (1) the intrusive
sheets continued to deform more easily during dyking compared
with their wall rocks and (2) the wall rocks were heated during
repeated granite sheeting. The latter point is illustrated by the
lack of chilled margins in intrusive granite sheets and the
presence of pervasive, high-temperature solid-state deformation
fabrics. Both features are difficult to reconcile with the shallow
level of granite emplacement proposed by previous workers
without invoking the localized heating of the wall rocks.
The role of synmagmatic deformation
Magma emplacement and deformation along the western margin
of the Mpuluzi batholith was controlled by the synmagmatic
Welverdiend shear zone. Significantly, both intrusive relationships and strain intensity in rocks of the GMS change abruptly
beyond the termination of the Welverdiend shear zone, underlining the significance of synmagmatic shearing for the emplacement of the Mpuluzi batholith as a whole. The spatial
distribution of coaxial shortening fabrics in the south and noncoaxial fabrics indicating dextral strike-slip shearing along the
northern extent of the Welverdiend shear zone suggests, in the
simplest scenario, a principal NW–SE-directed shortening strain
during regional deformation (Fig. 8). The clockwise rotation of
the foliation along the northern extent of the Welverdiend shear
zone is consistent with the swing of the foliation at, and close to,
the termination of a dextral strike-slip shear zone. The ESEtrending margin of the Mpuluzi batholith is, thus, located in the
extensional sector of the shear-zone termination. The dilational
component in this area that has created space for the granites is
evidenced by the ‘passive’ style of emplacement, the largely
undeformed nature of the granites and the lack of wall-rock
strains adjacent to the batholith. Similarly, both the location and
orientation of the Weergevonden tail (Figs 2 and 8) correspond
to an emplacement of the syeno-granites into, for example, an
extensional horsetail or a normal fault at the termination of the
Welverdiend shear zone. The NW–SE trend of the dyke-like
Weergevonden tail and the Kees Zyn Doorns syenite is also
consistent with their emplacement during regional NW–SEdirected shortening.
The regional-scale extent of the D3 tectonism in the granite–
gneiss terrane south of the Barberton greenstone belt is indicated
by the synmagmatic deformation of the Mpuluzi batholith
described by Jackson & Robertson (1983) from the Motjane
schist belt (Fig. 1) some 30 km SE of the Welverdiend shear
zone. The two gneiss belts describe an en echelon arrangement,
and planar and linear fabric elements as well as the NW–SE
shortening strains are similar in the Welverdiend shear zone and
the eastern gneiss belt (Jackson & Robertson 1983). Jackson &
Robertson (1983) did not record kinematic indicators in their
early work, but given the similar structural inventory and timing
of the two gneiss belts, we find it reasonable to speculate that the
gneisses to the SE of the Welverdiend shear zone also record a
component of dextral transcurrent shear. In this scenario, the
bulk of the Mpuluzi batholith occupies a dilational jog bounded
by the two NE-trending synmagmatic shear zones (Fig. 8). The
ESE-trending, sharply discordant and unstrained northern margins of the Mpuluzi batholith, in contrast, attest to the intrusion
of the granites into the extensional sector of the dilational jog
(Fig. 8). This regional model of synmagmatic, NW-trending, en
echelon shear zones and the resulting dilational jog geometry is
able to reconcile the seemingly contrasting intrusive relationships
386
Anhaeusser, C.R. 1973. The evolution of the early Precambrian crust of southern
Africa. Philosophical Transactions of the Royal Society of London, Series A,
273, 359–388.
Anhaeusser, C.R. 1980. A geological investigation of the Archaean granite–
greenstone terrane south of the Boesmanskop syenite pluton, Barberton
Mountain Land. Transactions of the Geological Society of South Africa, 83,
93–106.
Anhaeusser, C.R. 2001. The anatomy of an extrusive–intrusive Archaean mafic–
ultramafic sequence: the Nelshoogte schist belt and Stolzburg Layered
Ultramafic Complex, Barberton greenstone belt, South Africa. South African
Journal of Geology, 104, 167–204.
References
This material is based upon work supported by the National Research
Foundation under grant number NRF 2053186. We greatly appreciate the
co-operation of all landowners in the area during our fieldwork and thank
C. Anhaeusser for sharing his regional expertise with us. J. Reavy and T.
Blenkinsop are thanked for helpful reviews.
factors are intrinsic to the magma, including magma composition
and viscosity, rate of heat loss during ascent, magma driving
pressure, and the supply rate. Other factors are intrinsic to the
wall rocks and include the lithostatic load, the magnitude and
orientation of regional tectonic stresses, and the presence and
orientation of mechanical anisotropies.
The intrusion of the Mpuluzi batholith occurred during NW–
SE-directed subhorizontal shortening. Under these conditions, ó1
and ó3 are likely be horizontal at depth, whereas the intermediate
principal stress, ó2 , is vertical. This agrees with the strike-slip
kinematics recorded along the Welverdiend shear zone if a ‘nearAndersonian’ behaviour of the bounding shear zone is assumed.
At shallower crustal levels and with a progressive decrease of the
vertical load of the rock column, the least compressive stress, ó3 ,
will be vertical, having swapped its orientation with ó2 . This
allows for the vertical dilation of the granite sheets. The depth at
which this transition of the intermediate and least compressive
stress occurs is sometimes referred to as the ‘critical depth’
(Brisbin 1986). The important consequence of the critical depth
for the propagation and orientation of the granitic sheets is
obvious. Subvertical granite sheets are favoured at depth,
whereas subhorizontal sheets will dominate above the critical
depth at shallow crustal levels. Hogan et al. (1998) have
discussed the shape and orientation of granite sheets as a
function of the relative magnitudes of magma driving pressure,
lithostatic load and the presence of subhorizontal strength
anisotropies in the wall rocks. The transition from subvertical
dykes to subhorizontal sheet-like bodies is commonly observed
to occur along subhorizontal strength anisotropies in the crust,
such as the brittle–ductile transition or lithological boundaries. A
prerequisite for the formation of subhorizontal granite sheets is
that the magma driving pressure is sufficiently large to lift the
overburden. The roof rocks of the Mpuluzi batholith are nowhere
exposed, so that one can only speculate about possible rheological and/or mechanical controls of the overlying wall rocks on the
emplacement and lateral spreading of the granites. Wall-rock
xenoliths in the Mpuluzi batholith indicate, however, that the
GMS suite intruded into banded TTG gneisses and greenstones
similar to the flanking wall rocks. Given the mainly steep dips of
the basement gneisses, the strength anisotropy of the wall rocks
had evidently no control on the emplacement of the subhorizontal sheets of the Mpuluzi batholith. In the absence of any other
obvious controls, we suggest that the transition from predominantly subvertical sheets to the subhorizontal tabular geometry of
the Mpuluzi batholith tracks the location of the critical depth at
the time of intrusion.
levels below the main mass of the subhorizontal Mpuluzi granite
sheet. Similar intrusive relationships to those along the escarpment are exposed in deeply incised river sections that cut
laterally for several kilometres into the central portions of the
Mpuluzi batholith. The largest parts of the Mpuluzi batholith
appear to be underlain by stockworks or swarms of multiple
dyke- and sheet-like intrusions. Notably, granite dyking and
sheeting is not observed in the TTG basement outside the
confines of the batholith. Areas of pervasive granite sheeting are,
thus, confined to regions that underwent active, synmagmatic
deformation. This includes the bounding shear zones, such as the
Welverdiend shear zone, and extensional sectors at either the
shear-zone termination, or, on a broader scale, in dilational jogs
between bounding shear zones (Fig. 8). Repeated magma injection and sheeting is probably related to slip events along the
bounding Welverdiend shear zone and associated dilation. Granite sheeting into and parallel to active strike-slip shear zones
such as the Welverdiend shear zone and, thus, at high angles to
the bulk shortening strain is a widely documented feature
(Hutton 1992; Fowler 1994). The intrusion of the foliationparallel sheets probably tracks planes of weakness, that is, tensile
strength anisotropies represented by the foliation planes in the
developing shear zone (e.g. Hutton 1992). Given that large tracts
of the Welverdiend shear zone are oriented at high angles to the
regional NW–SE shortening strain, high magma pressures and
consequently low effective pressures within the shear zone have
also probably promoted transcurrent shearing along the Welverdiend shear zone. Sharply discordant granite dykes along the
northern margin of the Mpuluzi batholith show scattered but
predominantly ESE trends. These trends agree, at least within
20–258, with an emplacement of the dykes into extensional
fractures that opened during the NW–SE-directed regional shortening strain.
The overall intrusive pattern at lower structural levels is that of
a network of relatively small-scale, interlinked magma conduits
below the main, tabular Mpuluzi batholith. This network corresponds in many respects to the structurally controlled pervasive
magma transfer described by Collins & Sawyer (1996). A
difference is that magma transfer occurred through mainly
distinct granite sheets rather than along pervasive, mainly
dilational structures during ductile deformation as described by
Collins & Sawyer (1996). This may reflect the relatively shallow
levels of emplacement of the Mpuluzi batholith and the mainly
brittle behaviour of country rocks. Brittle fracturing and sheeting
probably occurred in the presence of high magma pressures and
high strain rates during sheet propagation, despite the fact that
wall rocks were undergoing ductile deformation during intrusion.
The intrusive features illustrated here for the Mpuluzi batholith seem to be applicable to other large batholiths in the region
such as the eastern parts of the Mpuluzi batholith (Hunter 1973)
and the Nelspruit batholith to the north (Robb et al. 1983) (Fig.
1). Notably, Robb et al. (1983) also described NE-trending,
laterally extensive migmatite–gneiss belts for the Nelspruit batholith, but the actual structural controls probably need to be
evaluated individually for each pluton.
The dyke–sheet transition
A striking feature of the Mpuluzi batholith and other batholiths
of the GMS suite in the region is the rapid transition from
subvertical dykes and sheets at deeper levels to subhorizontal
sheets at higher structural levels. The shape and orientation of
granitic sheets are determined by a variety of factors (Brisbin
1986; Hogan et al. 1998; Holdsworth et al. 1999). Some of these
387
Hutton, D.H.W. & Ingram, G.M. 1992. The Great Tonalite Sill of southeastern
Alaska and British Columbia: emplacement into an active contractional high
angle reverse shear zone. Transactions of the Royal Society of Edinburgh:
Earth Sciences, 83, 383–386.
Jackson, M.P.A. & Robertson, D.I. 1983. Regional implications of Early
Precambrian strains in the Onverwacht Group adjacent to the Lochiel granite,
north-west Swaziland. In: Anhaeusser, C.R. (ed.) Contributions to the
Geology of the Barberton Mountain Land. Special Publication of the
Geological Society of South Africa, 9, 45–62.
Jelsma, H.A., Van Der Beek, P.A. & Vinyu, M.L. 1993. Tectonic evolution of
the Bindure Shamva greenstone belt (northern Zimbabwe): progressive
deformation around diapiric batholiths. Journal of Structural Geology, 15,
163–176.
Kamo, S.L. & Davis, D.W. 1994. Reassessment of Archaean crustal development
in the Barberton Mountain Land, South Africa, based on U–Pb dating.
Tectonics, 13, 167–192.
Kisters, A.F.M., Stevens, G., Dziggel, A. & Armstrong, R.A. 2003.
Extensional detachment faulting and core-complex formation in the southern
Barberton granite–greenstone terrain, South Africa: evidence for a 3.2 Ga
orogenic collapse. Precambrian Research, 127, 355–378.
Krogh, T.E. 1982. Improved accuracy of U–Pb ages by the creation of more
concordant systems using an air abrasion technique. Geochimica et Cosmochimica Acta, 46, 617–649.
Ludwig, K.R. 1993. A Computer Program for Processing Pb–U–Th Isotope Data,
Version 1.24. US Geological Survey, Open-File Report, 88-542.
Ludwig, K.R. 2000. Isoplot/Ex: a Geochronological Toolkit for Microsoft Excel.
Berkeley Geochronology Center, Berkeley, CA.
Parrish, R.R. 1987. An improved micro-capsule for zircon dissolution in U–Pb
geochronology. Chemical Geology, 66, 99–102.
Paterson, S.R. & Fowler, T.K. Jr 1993. Re-examining pluton emplacement
mechanisms. Journal of Structural Geology, 15, 191–206.
Paterson, S.R., Vernon, R.H. & Tobisch, O.T. 1989. A review of criteria for the
identification of magmatic and tectonic foliations in granitoids. Journal of
Structural Geology, 11, 349–363.
Petford, N., Cruden, A.R., McCaffrey, K.J.W. & Vigneresse, J.-L. 2000.
Granite magma formation, transport and emplacement in the Earth’s crust.
Nature, 408, 660–673.
Pitcher, W.S. 1970. Ghost stratigraphy in granites: a review. In: Newall, G. &
Rast, N. (eds) Mechanisms of Igneous Intrusion. Geological Journal Special
Publication, 2, 123–140.
Poujol, M. & Anhaeusser, C.R. 2001. The Johannesburg Dome, South Africa:
new single zircon U–Pb isotopic evidence for early Archaean granite–
greenstone development within the central Kaapvaal Craton. Precambrian
Research, 108, 139–157.
Ramsay, J.G. 1989. Emplacement kinematics of a granite diapir: the Chinamora
batholith, Zimbabwe. Journal of Structural Geology, 11, 191–209.
Ridley, J.R., Vearncombe, J.R. & Jelsma, H.A. 1997. Relations between
greenstone belts and associated granitoids. In: De Wit, M.J. & Ashwal,
L.D. (eds) Greenstone Belts. Oxford University Press, Oxford, 377–397.
Robb, L.J. & Anhaeusser, C.R. 1983. Chemical and petrogenetic characteristics
of the Archaean tonalite–trondhjemite gneiss plutons in the Barberton
Mountain Land. In: Anhaeusser, C.R. (ed.) Contributions to the Geology of
the Barberton Mountain Land. Special Publication of the Geological Society
of South Africa, 9, 103–116.
Robb, L.J., Anhaeusser, C.R. & Van Nierop, D.A. 1983. The recognition of the
Nelspruit batholith north of the Barberton greenstone belt and its significance
in terms of Archaean crustal evolution. In: Anhaeusser, C.R. (ed.)
Contributions to the Geology of the Barberton Mountain Land. Special
Publication of the Geological Society of South Africa, 9, 117–130.
Stacey, J.S. & Kramers, J.D. 1975. Approximation of terrestrial lead isotope
evolution by a two stage model. Earth and Planetary Science Letters, 26,
207–221.
Stevens, G., Droop, G.T.R., Armstrong, R.A. & Anhaeusser, C.R. 2002. The
Schapenburg schist belt: amphibolite-facies metamorphism in Fig Tree Group
metasediments as a consequence of c. 3.23 Ga terrane accretion in the
Barberton greenstone belt, South Africa. South African Journal of Geology,
105, 271–284.
Tikoff, B. & Teyssier, C. 1992. Crustal-scale, en-echelon ‘P-shear’ tensional
bridges: a possible solution to the batholithic room problem. Geology, 20,
927–930.
Van Kranendonk, M.J., Collins, W.J., Hickman, A.H. & Pawley, M.J. 2004.
Critical tests of vertical vs. horizontal tectonic models for the East Pilbara
Granite–Greenstone Terrain, Pilbara Craton, Western Australia. Precambrian
Research, 131, 173–211.
Vauchez, A., Pacheco Neves, S. & Tommasi, A. 1997. Transcurrent shear zones
and magma emplacement in Neoproterozoic belts of Brazil. In: Bouchez,
J.L., Hutton, D.H.W. & Stephens, W.E. (eds) Granite: from Segregation of
Melt to Emplacement Fabrics. Kluwer Academic, Dordrecht, 275–294.
Anhaeusser, C.R. & Robb, L.J. 1980. Regional and detailed field and geochemical
studies of Archaean trondhjemite gneisses, migmatites and greenstone
xenoliths in the southern part of the Barberton Mountain Land, South Africa.
Precambrian Research, 11, 373–397.
Anhaeusser, C.R. & Robb, L.J. 1983. Geological and geochemical characteristics
of the Heerenveen and Mpuluzi batholiths south of the Barberton greenstone
belt and preliminary thoughts on their petrogenesis. In: Anhaeusser, C.R.
(ed.) Contributions to the Geology of the Barberton Mountain Land. Special
Anhaeusser, C.R., Robb, L.J. & Viljoen, M.J. 1981. Provisional geological map
of the Barberton greenstone belt and surrounding terrane, eastern Transvaal
and Swaziland, scale 1:250.000. Geological Society of South Africa,
Johannesburg.
Armstrong, R.A., Compston, W., De Wit, M.J. & Williams, I.S. 1990. The
stratigraphy of the 3.5–3.2 Ga Barberton greenstone belt revisited: a single
zircon ion microprobe study. Earth and Planetary Science Letters, 101,
90–106.
Brisbin, W.C. 1986. Mechanics of pegmatite intrusion. American Mineralogist, 71,
644–651.
Brown, M. & Solar, G.S. 1998. Granite ascent and emplacement during
contractional deformation in convergent orogens. Journal of Structural
Geology, 20, 1365–1393.
Clemens, J.D., Petford, N. & Mawer, C.K. 1997. Ascent mechanisms of granitic
magmas: causes and consequences. In: Holness, M.B. (ed.) Deformationenhanced Fluid Transport in the Earth’s Crust and Mantle. The Mineralogical
Series. Chapman & Hall, London, 144–171.
Collins, W.J. & Sawyer, E.W. 1996. Pervasive granitoid magma transfer through
the lower–middle crust during non-coaxial compressional deformation.
Journal of Metamorphic Geology, 14, 565–579.
Cruden, A.R. 1998. On the emplacement of tabular granites. Journal of the
Geological Society, London, 154, 853–862.
De Ronde, C.E.J. & De Wit, M.J. 1994. Tectonic history of the Barberton
greenstone belt, South Africa: 490 million years of Archaean crustal
evolution. Tectonics, 13, 983–1005.
De Wit, M.J., Armstrong, R.A., Hart, R.J. & Wilson, A.H. 1987. Felsic igneous
rocks within the 3.3 to 3.5 Ga Barberton greenstone belt: high crustal level
equivalents of the surrounding tonalite–trondhjemite terrain emplaced during
thrusting. Tectonics, 6, 529–549.
De Wit, M.J., Roering, C., Hart, R.J. & Armstrong, R.A. et al. 1992.
Formation of an Archaean continent. Nature, 357, 553–562.
Dubé, B., Dunning, G.R., Lauziere, K. & Roddick, J.C. 1996. New insights into
the Appalachian orogen from geology and geochronology along the Cape Ray
fault zone southwest Newfoundland. Geological Society of America Bulletin,
108, 101–116.
Dziggel, A., Stevens, G., Poujol, M., Anhaeusser, C.R. & Armstrong, R.A.
2002. Metamorphism of the granite–greenstone terrane south of the
Barberton greenstone belt, South Africa: an insight into the tectonothermal
evolution of the ‘lower’ portions of the Onverwacht Group. Precambrian
Research, 114, 221–247.
Fowler, T.J. 1994. Sheeted and bulbous pluton intrusion mechanisms of a small
granitoid from southeastern Australia: implications for dyke-to-pluton transformation during emplacement. Tectonophysics, 234, 197–215.
Frost, B.R., Chamberlain, K.R. & Schuhmacher, J.C. 2000. Sphene (titanite):
phase relations and role as geochronometer. Chemical Geology, 172,
131–148.
Guineberteau, B., Bouchez, J.-L. & Vigneresse, J.-L. 1987. The Mortagne
granite pluton (France) emplaced by pull-apart along a shear zone: structural
and gravimetric arguments and regional implication. Geological Society of
America Bulletin, 90, 763–770.
Hogan, J.P., Price, J.D. & Gilbert, M.C. 1998. Magma traps and driving
pressure: consequences for pluton shape and emplacement in an extensional
regime. Journal of Structural Geology, 20, 1155–1168.
Holdsworth, R.E., McErlean, M.A. & Strachan, R.A. 1999. The influence of
country-rock structural architecture during pluton emplacement: the Loch
Loyal syenites, Scotland. Journal of the Geological Society, London, 156,
163–175.
Hunter, D.R. 1957. The geology, petrology and classification of the Swaziland
granites and gneisses. Transactions of the Geological Society of South Africa,
60, 85–125.
Hunter, D.R. 1973. The granitic rocks of the Precambrian in Swaziland. In:
Lister, L.A. (ed.) Symposium on granites, gneisses and related rocks. Special
Hutton, D.H.W. 1988. Igneous emplacement in a shear-zone termination: the
biotite granite at Strontian, Scotland. Geological Society of America Bulletin,
100, 1392–1399.
Hutton, D.H.W. 1992. Granite sheeted complexes: evidence for the dyking ascent
mechanism. Transactions of the Royal Society of Edinburgh: Earth Sciences,
83, 377–382.
388
evolution of the Barberton Mountain Land, South Africa. PhD thesis,
Kingston University.
Zulauf, G. & Helferich, S. 1997. Strain and strain rate in a synkinematic
trondhjemite dike: evidence for melt induced strain softening during shearing
(Bohemian Massif, Czech Republic). Journal of Structural Geology, 19,
639–652.
Vigneresse, J.L. & Tikoff, B. 1999. Strain partitioning during partial melting and
crystallizing felsic magmas. Tectonophysics, 312, 117–132.
Viljoen, M.J. & Viljoen, R.P. 1969. An introduction to the geology of the
Barberton granite–greenstone terrain. In: Haughton, S.H. (ed.) Upper Mantle
Project. Special Publication of the Geological Society of South Africa, 2, 9–28.
Yearron, L.M. 2003. Archaean granite petrogenesis and implications for the
Received 27 February 2004; revised typescript accepted 19 August 2004.
Scientific editing by Rob Strachan

Field trip to the Mesoarchaean Barberton Granite-Greenstone

Transcription

Similar documents

Swallow Rocks to Big Mountain

Greenstone - Miami University

Brittany fieldtrip Easter 2013 First stop: Port Navalo, south Brittany

Yg SOC SO MD D M PIP Pa Kmb Tps Ti Tual Tla Tlrp Turf QTg QTp

05-16 FMA newsletter

Tactical Vest

zorritas

Crowd Control Signature Series

Boiling With Hot Stones